Record minutes of business meetings, workpackages meetings, and project meetings.
The first review of MOLTO takes place in Luxemburg on 15 March 2011.
Here are the slides.
Title | MOLTO Overview - review presentation 2011 |
Publication Type | Slide Presentation |
Authors | Ranta, A |
Year of Publication | 2011 |
Date Published | 03/2011 |
Publication Language | English |
Type of Work | Review slides |
Attachment | Size |
---|---|
intro-review-2011.pdf | 394.63 KB |
Title | WP1 - review presentation 2011 |
Publication Type | Slide Presentation |
Authors | Caprotti, O, Ranta, A |
Place Published | Luxemburg |
Year of Publication | 2011 |
Date Published | 03/2011 |
Publication Language | English |
Keywords | Review March 2011, WP1 |
Type of Work | Review slides |
Attachment | Size |
---|---|
review_MOLTO_WP1.pdf | 576.97 KB |
Title | WP2 - review presentation 2011 |
Publication Type | Slide Presentation |
Authors | Ranta, A |
Year of Publication | 2011 |
Date Published | 03/2011 |
Publication Language | English |
Type of Work | Review slides |
Attachment | Size |
---|---|
grammarians-tools.pdf | 765.61 KB |
Title | WP3, WP9 - review presentation 2011 |
Publication Type | Slide Presentation |
Authors | Carlson, L |
Year of Publication | 2011 |
Date Published | 03/2011 |
Publication Language | English |
Type of Work | Review slides |
URL | http://www.helsinki.fi/~lcarlson/MOLTO/MOLTO_review_UHEL_2011-03-15.pdf |
Attachment | Size |
---|---|
MOLTO_review_UHEL_2011-03-15.pdf | 469.39 KB |
Title | WP4 - review presentation 2011 |
Publication Type | Slide Presentation |
Authors | Popov, B, Mitankin, P |
Year of Publication | 2011 |
Date Published | 03/2011 |
Publication Language | English |
Type of Work | Review slides |
Attachment | Size |
---|---|
WP4.Y1.pdf | 323.56 KB |
Title | WP5 - review presentation 2011 |
Publication Type | Slide Presentation |
Authors | España-Bonet, C, Màrquez, L |
Year of Publication | 2011 |
Publication Language | English |
Keywords | Review March 2011, WP5 |
Attachment | Size |
---|---|
WP5_mar11.pdf | 452.4 KB |
Title | WP6 - review presentation 2011 |
Publication Type | Slide Presentation |
Authors | Saludes, J, Xambó, S |
Year of Publication | 2011 |
Date Published | 03/2011 |
Publication Language | English |
Keywords | Review March 2011, WP6 |
Type of Work | Review slides |
Attachment | Size |
---|---|
wp6-y1.pdf | 330.05 KB |
Title | WP7 - review presentation 2011 |
Publication Type | Slide Presentation |
Authors | España-Bonet, C, Màrquez, L |
Year of Publication | 2011 |
Publication Language | English |
Keywords | Review March 2011, WP7 |
Attachment | Size |
---|---|
WP7_mar11.pdf | 553.81 KB |
Title | WP10 - review presentation 2011 |
Publication Type | Slide Presentation |
Authors | Caprotti, O |
Place Published | Luxemburg |
Year of Publication | 2011 |
Date Published | 03/2011 |
Publication Language | English |
Keywords | Review March 2011, WP10 |
Type of Work | Review slides |
Attachment | Size |
---|---|
review_MOLTO_WP10.pdf | 785.58 KB |
Collection of slides from the KickOff Meeting, Barcelona 2010 at
http://www.molto-project.eu/category/publication/molto-dissemination/mol...
Time Slot | Title | Speaker |
---|---|---|
9:35 | MOLTO Introduction | Aarne |
10:00 | WP1 management report | Olga |
10:10 | WP2 grammar tools | Aarne |
cloud demo WP2, WP3 | Aarne | |
WP3 translators tools, (lexicon tools) | Inari | |
term factory demo WP3, WP4 | Lauri | |
WP4 grammar/ontology (query language) | Maria | |
query demo WP4, WP7, WP8 | Maria | |
WP5 robust and hybrid translation | Cristina | |
WP6 mathematics | Jordi | |
WP7 patents | Txell | |
WP8 museum | Aarne | |
WP9 evaluation | Lauri | |
WP10 dissemination exploitation | Olga, Borislav, Joris | |
WP11 AceWiki-GF | Kaarel | |
WP12 business application | Joris |
List of participants: Olga Aarne Inari Lauri Maria Cristina Jordi Borislav Joris Kaarel Txell
Testing the slides directly from here.
A summary of the boring things.
Please add a child page for your workgroup.
The working groups will meet on January 13, 2012. You can probably also enable signup for a workgroup. My suggestion is that you take this as an opportunity to create Task descriptions (they will be automatically included in the work package page) and link and log the group work here for future record.
Here is the log of the meeting.
Participants list:
Minutes and notes from the working groups day.
Participants: Ramona, Inari, Milen, LauriC
The task is to verbalize an ontology. The ontology in this case is the Gothenburg City Museum ontology, which contains information about the museum objects in that particular museum. We don't need to prepare for unrestricted vocabulary or user input; at least not in text form.
The GF grammar is two-part. We have a direct verbalization, that is, just the ontology triples translated into GF syntax. As one triple contains only one piece of information, the resulting sentences from this GF grammar are of type Mona Lisa is a Painting. Mona Lisa is painted by Leonardo da Vinci. Mona Lisa is located in Louvre. This grammar is still useful as a lexicon and facts about the items in the lexicon. Creating a concrete syntax needs necessarily some human work: the ontology does not contain translations for the terms in all languages we'd want, and even for languages it covers, it doesn't have all linguistic information needed in a GF grammar. (WP2 and WP3 tools could be used to help the grammar writer; more on that later.)
In addition to the direct verbalization, we have also a discourse-building grammar. It consists of GenText.gf
and GenLexicon.gf
. The GenLexicon is a more refined version of the direct verbalization grammar: it has some irrelevant parts removed (such as numerical codes, which have no information value for a human) and some of the lexical information already aggregated. For example, the ontology triples for a painting might look like this:
xx:Guernica yy:paintedBy zz:Picasso xx:Guernica cc:hasColour dd:Black xx:Guernica cc:hasColour dd:White xx:Guernica cc:hasColour dd:Gray
We can see that there are multiple instances of the predicate hasColour
, and we can aggregate that information into only one colour in the GF grammar, called BlackWhiteGray
.
(A more generic approach would be like this:
White : Colour ; Black : Colour ; Red : Colour ; Multicolour : Colour -> Colour -> Colour ; Multicolour Red (Multicolour Black White) Multicolour Black (Multicolour Red (Multicolour White Black))
We could generate every combination of colours with that, but as the task is to verbalize only the paintings in Gothenburg Museum Ontology, we can get every existing combination and have them in the grammar spelled out. This means more items in the GF grammar, but it makes the language more fluent: many languages have specific words for some common combinations of colours, but with the Multicolour
way of combining colours we'd have to choose a generic rule with which to combine colours, such as black and white and grey and blue and red.)
Finally, GenText.gf
is the grammar where we do the information aggregation on a semantic level. We build multiple discourse patterns; since there are many ways to give information about a painting. This is in practice implemented with huge bunch of dependent types: a sentence that tells a painting's author, colour and location is of type Painting Author Location
, and the items are all checked to be compatible with the ontology; we can't build a sentence "Guernica was painted by Pablo Picasso and it is located in Helsinki" if the details don't match with the direct verbalization grammar. We can create proof objects automatically for every original triple. There are also proof objects with 3 or more arguments (and they are created by ???).
Anyway, the functions in GenText look somewhat like this:
BuildSentence -> (p: Painting) -> (a: Author) -> (l: Location) -> AuthorProofObject p a -> LocationProofObject p l -> FinalProofObject p a l -> Text ;
We want translations of the terms, not new ones. We can use WP3 lexicon management tools to search FactForge's art terms. We can also use the example-based grammar writing tool.
Allow users to query the ontology and output exactly the answers they want. But that's not for the next deliverable.
Participants: Ramona, LauriC, LauriA, Aarne, Jordi, John, Inari, everyone
Mod Boring Pizza
Kind
is a noun phrase without article, and if we ask how to say Boring Pizza
, a user might add an article, and then the user input can't be parsed as a CN
Boring Fish
is "boring fish", then Boring Cheese
is "boring cheese".Participants: Aarne, Ramona, Milen, Borislav, Cristina, Meritxell
Write GF grammars to solve problems in the SMT system: compound and biological names, word reordering, gender agreement.
Biological names are different than compound names and raise different challenges. LauriC can provide a database of compounds and biological names.
Increase parser robustness by chunking the claims and parsing the chunks separately and recombine the results with the help of the grammar. Reduce ambiguities with bottom up disambiguation based on the corpus.
Simplify the query language because some of the English queries are weird in other languages.
Will need someone else to work on French grammars.
The user interface allows querying the retrieval system using the controlled language, free text and a combination of both.
Results are shown highlighting the relevant words in relation to the query. If not possible to find the words in translated documents (lexicon is needed) then highlight the whole sentence.
GF and SMT tools have been shared and installed to start hybridisation studies.
Generate synthetic corpora and alignments from GF grammars.
Provide a translation of the lexicon.
Write simple abstract syntax representation and grammar for the results. Write templates for each topic.
Comparison between equivalent tools in GF and SMT systems.
This WG is the same as Hybrid Systems & Patents WG .
Participants: Lauri A, Lauri C, Chunxiang, Inari
Participants:
Notes taken during the MOLTO Kickoff meeting
Add your own page
Upcoming deliverables -M6
Dissemination upcoming
Deputy for project manager OC to be decided.
The above choices are considered final unless email to the contrary with nomination by 17 March 2010. Final composition will be online by March 20, 2010.
1st project meeting: - to finalize the progress report: M7=September, M13=March - Ontotext will host 1st Project Meeting, possibly Varna / Sofia - public program (1 day) and internal meeting (2 day) - prefer 2nd week - JS says courses start mid September - 6-12 Sept.2010
OC: to setup a poll with dates and one with location.
2nd project meeting: - March 2011: UGOT (we keep 2nd week of March because has no courses in UHEL and UGOT)
3rd project meeting: - September 2011: UHEL
4th project meeting: - March 2012: MXW
5th project meeting: - September 2012: UPC (24th Sept. is La Mercè)
6th project meeting: - March 2013, FINAL meeting at UGOT
D9.1 needs input also from WPs that will start later - e.g. the cultural heritage case study. In this case (WP8), we envision providing specific criteria at beginning of WP in addition to the general criteria submitted in D9.1.
Website
- logo poll: 100% votes YES to keep logo (papyrus font).
- partners logo - to be included in description pages of partners on molto website.
- background - make it lighter or remove it.
- add google analytics (grant read access to google analytics)
Aspects that are relevant for a Grant Agreement (not a contract as in FP6).
Definition of the GA:
- Core part
- Annex II
- Annex I
Commission must be informed about _any_ major events
Communication in writing - event. amendments
### Amendments
can come from commission too
### Reporting
via NEFF? system online (same used in the negotiation)
- within 60 days after end of rep. period
- fill-in all fields in template in annex VI
- if _cumulated_ costs > 375kEur - use form D annex VI to get audit
- cumulative summary after each period
__ Checks __
- Euro exchange rate: official rate
- right signatures?
- dated and signed?
__ Settlement __
- 105 days to pay
### Reporting
- Annual progress report
- Annual report on resources usage
- Annual financial statement
- 6 months interim Progress report
- Public summary in the progress reports
### Eligible costs
*not-eligible*
- taxes incl. VAT
- non-productive time of researchers
Keep records up to 5 years.
### Certificate on the methodology
average personnel costs and overhead - on a voluntary basis - must be accepted by EC
Please enter your notes here is you have any.
Thanks.
Attachment | Size |
---|---|
GPF-1.pdf | 4.01 KB |
Location Sal de Llac, UPC, Barcelona
9:00
12:30 lunch
14:00
15:00 demos
16:00 feedback from EC and reviewers: Michel, Vincent, Vangelis
About 15-20 minutes per WP, incl. discussions.
SLIDES of presentations (pls upload and tag with review tag): http://www.molto-project.eu/category/publication/wp1/review-march-2012
Tuesday, 15 June - 15:00 CEST.
Present: AR, LC, BP, JS, OC (late)
Brief status report :
UGOT: Deliverables finished: dissemination plan, web service. Work started on GF-SMT hybrid in collaboration with UPC, first experimenting with alignments. Java-PGF interpreter up and running on Android phones. Krasimir has worked in Sofia and Varna and discussed with Ontotext.
UHEL: Started implementing a platform for evaluation. Getting into GF development, contributing to web service demo. Ontology-lexicon interface will soon be demoed in connection to the Content Factory project.
UPC: Experiments with OpenMT platform, which will be transferred to GF-SMT hybrid. First experiments in GF-SMT by Cristina with Ramona at UGOT. Mathematics case study: data collection started. Python bindings for GF, work towards integration with NLTK.
Ontotext: RDF infrastructure in place, with default knowledge bases; waiting for real data from other partners. Specification of query language sent to UGOT, who is expected to build a GF grammar. Looking at the extraction of ontology from Museum database obtained from Gothenburg City Museum via Dana at UGOT.
Finalize dates of 1st project meeting :
8-10 September in Varna.
Discuss program & logistics of 1st project meetings :
Day 3: technical work in groups around common tasks
Ontotext will book venue and give recommendations on accommodation and travel.
Upcoming deliverables in September :
Deliverables from UHEL are under way - no special action needed.
Other matters :
;Present: AR, OC, LC, JS, PM
UGOT: patent case study partner - nothing decided yet - ongoing discussing with EPO, AR is waiting reply from EPO
Kiril Simov has not replied for Varna - so no invited speakers so far
AR and OC will prepare the progress report to be completed in Varna by partners
AR: do we have internet in Varna?
PM: not sure about the internet situation in Varna - will bring router to Varna
AR: phrasebook led to experiments and ideas for tools, work is proceeding as expected
OC: ACL2010 meeting went well - AR, maybe LREC better venue for MOLTO
UHEL: LC put a draft on the wiki, difficulty is in getting info from users, afraid that WP are so independent until very end, and user testing gets done only in the end -- time scheduling to be reviewed in Varna (1 hour time in varna minimum to discuss)
UHEL wiki is open as sketchpad
UHEL is winding up another project and produced a tool for multilingual ontologies - will present it in Varna (project is ConFactory) - Inari and LauriA will probably work in a group with Petar on Ontologies
UPC: SMT group, cristina asked access to carry out patent study - but very slow
WP6, started with modularize webalt grammars, stuck now on german, naive approach: add new language by changing the lexicon and seeing if it compiles -- no idea what will happens with languages such as bul, fin, etc.
AR: let's get few languages in high quality first, then scale up
JS: eng spa, cat fre, ita, in the OpenMath layer compile
JS: cristina and ramona worked on integration run some experiments to create statistical model from GF generated parallel corpora -- also made word alignment in GF
JS: python plugin for GF in contrib folder -- it is a shared library can be imported in GF so one can open a pgf file from python to check grammars, linearize
expect to be difficult to integrate into main python core NLTK - easier to get into contrib
Onto: onto worked on 1. knowledge rep infrastructure 2. grammar ontology interoperability - demo planned for varna for #2
AR made GF query grammar - linked to ontology of ontotext so that natural language queries are executed by semantic engine
UI will be shown in Varna, uses autocompletion from gf
some efficiency problem being discussed with Krasimir, most likely our virtual machine makes things slower
AR: is query grammar enough or needs extension?
PM: maybe another case study can prove the point of ontology-grammar interoperability: lexicon is kept separate from grammar - it is generated from the ontology
PM will ask Simov tomorrow
Cristina might not be there on 1st day (amend program)
LC could get some material on evaluation of MT systems but not really a tutorial
Day 3 WGs :
Logistics: maybe 3 rooms for day 3 and larger room for day 1
Deliverables last done is D9.1 which is not stable - partners are supposed to continue work in evaluation
WP7 delayed 6 months, Cristina leads it. Deliverables pushed only 3 months - a milestone pushed also. Aarne partitioned the budget and looks ok so far but wait the EU project officer to open the online platform. OC: to update the deliverables online OC: upload DoW (under revision version - but budget summary not in yet) AA: Consortium agreement will be amended
Georg Artelsmair from EPO "fit for purpose" translations are ok, cooperate with Pluto, with Google too. We get3 the data from EPO, not try to compete with Google or Pluto. Suggestion to contact eg Swedish Patent Office or other national patents office. Ask them for specific tasks targeting minor languages. From Meta-Forum: clear that translation is a hot topic. MOLTO has to profile itself differently. BP: we can work together instead of competing. AR: yes, our hybrid model does that, we also need to focus on technology and in retrieval. AR: corpus still possibly pharmaceutical/chemical. EPO suggested "telecommunication" for northern countries. Unclear who will evaluate (for TAUS? we need to subcontract?) It will become clear once we start. LC: ECA in Helsinki has done chemistry eval before. AR: new call for SME - deadline is April for short 5 page proposals
school.grammaticalframework.org
XSD: in Spain this week is a holiday, which explains slow answers from us.
AR: we could add a new case study - ideal partner: would do a MOLTO application plus evaluation. OC: prepare call for partners. OC: will check money for this obj. XSD: new partners should strengthen MOLTO position in the translation arena
You may be aware that the 2011-2012 work programme includes Objective ICT 2011.11.3 - Supplements to strengthen co-operation in ICT R&D in an enlarged European Union. This offers to co-ordinators of ongoing ICT research projects (IP or STREP) which finish after the 30th June 2012 the opportunity to add one or more new partners to their project to expand its activities in or into the enlarged European Union.
Full details of how to submit your proposal are available on cordis.
Please go to the Guide for applicants for either IP's or STREPS and have a look at Annex 7 (p.63) which gives particular instructions for Objective 11.3.
To express your interest in finding new partners please make an entry.
The funding requested should represent a reasonable extension of the on-going project to achieve the goals of the objective; not exceeding 30% of the Commission funding of the existing project or EUR 1 million, whichever is the lower.
Present: AR, OC, JS, PM, LC
Matrixware situation has been finalized with the revision of the budget and of the DoW (sent today to the EU). New documents will be made available to partners by OC.
UHEL has a new MOLTO person (Shen).
Demos for review can be more technical than those planned for the 2nd project meeting. Review is scheduled for 14.03. List of possible demos:
Prel. number of participants: 4 (UHEL) + 4 (UPC) + 2 (Onto) + 2 AB
OC: to distribute the forms for the financial reports to partners
WP4: new version of grammars produces too many nonsensical queries, server speed for KRI still problematic, maybe because of virtualization issues
(UHEL is testing the bilingual lexicon in wordnet format for KRI, FI-EN, 200000 senses --- as soon as testing is over, this will be given to Ontotext)
Confirmed: AR, OC, JS, BP, LC
these talks will be somewhat with mixed contributions from various sites. This to be finalized during meeting WP day.
WP6: evaluation to remove errors can now go via the webpage. OC: to ask TH to add a feedback page to demo. JS: ask Ramona to commit her fixes on Romanian RG.
WP4: ok, 2 languages: EN and SE. However the set of possible questions that can be asked is still limited and KB is not rich enough in the facts interrelations - so demo not too exciting. It has been evaluated only internally.
LC: to start thinking about evaluation strategy for WP4 and WP6 so we can many short evaluation cycles. First milestones could be the meeting (a bit too close probably).
Participants: OC, AR, LC, JS, BP, SN
AR: extension of MOLTO (great review points) but now there are some issues and extension is not so sure. We should know in the first days of May. Extension would mean 2 new case studies. Adam Slaski was extended to work 4 months, Adam and Ramon worked on the grammar for the patent case study, they will present on Tuesday. You can join by Skype. AR contacted Patrizia Biani and reply was hopeful that the corpus will be made available soon. LC: we decided people would concentrate on text, leaving chemical compound names aside. Is that what is happening? AR: yes, this is what is happening. LC: good because we have an interest in that wrt to lexicon. AR: TH has developed further the editor, next version will have improved server connection and better feedback. I myself worked on WP6 related issues, in rendering logical connectives. WP8 work on grammars has also started but I have not seen any concrete result yet.
LC: we had few demos of the term factory system and we are moving it to the uni server so it will be available to the public. Combination with mediawiki and semantic wiki is interesting for our contact with Lea Laitinen (?) from a project funded by the Academy of Finland. Term factory basic tools are embeddable in platform like mediawiki, and we are planning to try it on Drupal. Full Ontology editor for the MOLTO translation tool is ready and will be integrated as lexicon editor.
BP: finishing deliverable on data models and alignments, with aspects related to the museum use case, the EN-FI wordnet, and also more general aspects. It is a document. Afterward we will load the same mappings, once we get the new server going. In the end we will have 1 repository with all the data sets loaded. We worked on the query infrastructure on top of DBpedia. PM is now in Academia, he comes to meet once a month, and we have now several people working on MOLTO. We are trying to recruit a new person with a different profile.
OC: remind to BP and LC that there are revision of deliverables due, D4.1 and D9.1. Please take a look.
JS: cannot report about status of WP5 but we have new recruit. XSD has stopped because of his work load, he cannot work day to day. MG, Aris is new person working on WP6, operations layers is going on. OM layer is working and we need to test and evaluate it. Current Treebank is 140 productions in English, local experts have said it is ok. We have to add all other languages. Main task for Aris is to fix bugs - can you guys (local experts) fix your own bugs in the meantime?
LC: WP9 has to define how to do the testing and debugging of WP6 layer done.
AR: treebank is a great starting point. we should think of a tool allowing to annotate the translations. we could start with few languages to define the methodology and then scale up.
JS: for deadline 1st June, we do 4 languages in our core.
OC: MOLTO project will be at EAMT 2011, 30-31 May 2011.
LC: we suggest 31 May 2011. 26 May 2011 in Helsinki - try to agree.
MOLTO extension is probable but not sure yet.
Date was 31 Aug - 2 Sep. 2011. This is now fixed and UHEL will make reservations. If extension goes thru we will have 2 new nodes. Invited speakers should be decided. UHEL will contact several groups and people for the Open Day.
Participants: (LC), JS, AR, OC
We keep it short because of skype issues of Lauri.
JS: all 10 10 languages compile for WP6 del. Thought due 30 Jun. Have treebank of 140 prods of the OM layers in 3 languages being corrected by humans (spa, ger, eng) - more than 100 tickets for these: too much. Lack of specificity in the RGL cause lots of tickets, marked as RGL. List of issues is not public available- JS will extract tickets and publish them on web. Should we release now?
AR: yes, I think you should do the release and plan to correct the issues asap. Then as public release we can hope to be helped in bug fixing.
AR: Cristina takes over WP5, Meritxell González will do WP7 instead of Cristina
LC able to join - report on meeting - first Ramona talked about the museum wp - 2nd day I got help with GU tools and it is shaping up but i still have issues, TH is helping. Inari is going to Hungary. She would like to get the demo on Android and will get a phone or a tablet if necessary for the demo.
AR: easy to install on android market: phrasedroid.
LC: make suggestions for invited speakers, local hero suggestions Harri Arnola, SUNDA, (RB translation) and Kimmo Koskenniemi, Dept. Modern Languages, UHEL. ProMT company in Russia also a possibility.
AR: we have no Russian in GF now, maybe better contact them when we have it later on. We will have to do publicity.
[11/9/12 1:05:29 PM] Olga Caprotti added Aarne Ranta to this conversation [11/9/12 1:06:08 PM] Olga Caprotti added Jordi Saludes, Joris van Aart, Lauri Carlson, nefuchs to this conversation [11/9/12 1:06:25 PM] Olga Caprotti added Mateva to this conversation [11/9/12 1:14:57 PM] Olga Caprotti: hello [11/9/12 1:15:21 PM] Jordi Saludes: hello [11/9/12 1:15:23 PM] Mateva: hi Olga :) [11/9/12 1:15:25 PM] Olga Caprotti: Agenda ====== - status report - closing the progress report - deliverables delays and issues - requests to partners - dissemination - other items [11/9/12 1:15:36 PM] nefuchs: Norbert is here [11/9/12 1:15:37 PM] Joris van Aart: Hi all [11/9/12 1:15:52 PM] Mateva added borislav popov to this conversation [11/9/12 1:15:53 PM] Olga Caprotti: aarne, if you could issue the call from the dept [11/9/12 1:16:39 PM] Olga Caprotti: Aarne? [11/9/12 1:16:56 PM] Aarne Ranta: I dont have all skype addresses [11/9/12 1:17:01 PM] Olga Caprotti: we are all in here [11/9/12 1:17:30 PM] Aarne Ranta: This message has been removed. [11/9/12 1:17:52 PM] Olga Caprotti: Olga Caprotti has ejected borislav popov from this conversation [11/9/12 1:18:37 PM] Mateva: sorry my skype was disconnected [11/9/12 1:19:50 PM] Aarne Ranta: Maria can you come back [11/9/12 1:19:59 PM] Olga Caprotti: maria, you have a do not disturb on [11/9/12 1:20:49 PM] Mateva: I heard you [11/9/12 1:21:04 PM] Aarne Ranta: This message has been removed. [11/9/12 1:21:08 PM] Olga Caprotti: aarne, it is easier if you hang and call again [11/9/12 1:21:22 PM] Aarne Ranta: This message has been removed. [11/9/12 1:21:54 PM] Olga Caprotti: Agenda ====== - status report - closing the progress report - deliverables delays and issues - requests to partners - dissemination - other items [11/9/12 1:24:21 PM] Olga Caprotti: Participants: MM, AR, JS, JvA, NEF [11/9/12 1:24:25 PM] Olga Caprotti: Missing: LC [11/9/12 1:25:49 PM] Olga Caprotti: MM: need on improvements in patents prototype WP7, started work on WP8, on KB - have reloaded it and worked on website -- trying to expose answers in NL for 7 or 8 languages. Added one language for RDF to GF grammar [11/9/12 1:26:59 PM] Olga Caprotti: Mariana@Ontotext is doing some research related to WP8, Ilias helped with annotation of the the patents documents in the prototype [11/9/12 1:28:30 PM] Olga Caprotti: annotation pipeline was improved by other people too -- to expose more of them. mostly technical work, and research on GF-Lemon approach [11/9/12 1:29:04 PM] Olga Caprotti: Ontotext is on schedule wrt ongoing work. [11/9/12 1:30:01 PM] Olga Caprotti: NEF: we are considering the museum example as use case example for AceWIki GF - exp. date is January [11/9/12 1:30:15 PM] Olga Caprotti: MM: I need the query grammar in order to proceed [11/9/12 1:30:29 PM] Olga Caprotti: AR: expect the query grammar to be done in 2 weeks [11/9/12 1:30:35 PM] Mateva: borislav can be added now [11/9/12 1:31:16 PM] Olga Caprotti: NEF: we shall take Jan. to setup the framework. [11/9/12 1:31:38 PM] Olga Caprotti: AR: do you want sparql answers too? [11/9/12 1:31:46 PM] Olga Caprotti: NEF: we want answers in NL [11/9/12 1:32:13 PM] Olga Caprotti: OC: Ontotext, UZH, UGOT to get WP8 running [11/9/12 1:32:52 PM] Olga Caprotti added borislav popov to this conversation [11/9/12 1:34:09 PM] Olga Caprotti: NEF: add gf grammar directly to acewiki, will be available public soon. for evaluation, use cases Ace in GF, MOLTO phrasebook or the museum example. Ace in GF now in 6 languages. still some grammar problems to be polished, hope JC works on this. [11/9/12 1:34:29 PM] Olga Caprotti: NEF: we need to send grammar for evaluation by end of jan. [11/9/12 1:34:41 PM] Olga Caprotti: NEF, we can make next deliverable. [11/9/12 1:34:56 PM] Olga Caprotti: NEF: i am implementing arithmetics in CLP. [11/9/12 1:35:08 PM] Olga Caprotti: NEF: this is my work in race [11/9/12 1:35:29 PM] Olga Caprotti: NEF: we need input on what you expect concerning evaluation. [11/9/12 1:35:54 PM] Olga Caprotti: NEF: we ask for suggestion. what will UHEL do with our grammar? [11/9/12 1:36:20 PM] Olga Caprotti: NEF: we wonder how the usability of Acewiki will be tested. [11/9/12 1:36:35 PM] Olga Caprotti: AR: will press UHEL on this. [11/9/12 1:36:54 PM] Olga Caprotti: NEF: maybe you can setup a group of student to test Acewiki [11/9/12 1:37:20 PM] Lauri Carlson: Hello from Lauri C [11/9/12 1:38:15 PM] Olga Caprotti: NEF: personnel problem, Tobias Kuhn has applied for a job and he is expected to start jan1. we tried for a compromise, he is crucial, we do not want tobias career to be damaged. [11/9/12 1:39:07 PM] Olga Caprotti: NEF: i was invited to a scientific meeting, with 6,7 people, mid november. will report on MOLTO. [11/9/12 1:39:24 PM] Olga Caprotti: NEF: a MOLTO dissemination activity [11/9/12 1:40:02 PM] Aarne Ranta added lauri.carlson to this conversation [11/9/12 1:40:17 PM] Olga Caprotti: JS: WP5 trying to homogeneize eng->fr in eng->de but not really working - waiting for package from UGOT [11/9/12 1:40:24 PM] Aarne Ranta: welcome Lauri [11/9/12 1:40:41 PM] Olga Caprotti: JS: in the students dialog, implementing the feedback grammar and the reasoner. [11/9/12 1:41:17 PM] Olga Caprotti: JS: need some help with GF grammar fine points, to make language fluent. Posted bugs in trac system [11/9/12 1:41:44 PM] Olga Caprotti: AR: trac must be configured to send email to relevant people [11/9/12 1:42:18 PM] Olga Caprotti: AR: WP6 abandoning integration with acewiki - running out of time. [11/9/12 1:43:02 PM] Aarne Ranta: last AR was JS [11/9/12 1:43:29 PM] Olga Caprotti: JS: optimizing from 1 week to 1 day [11/9/12 1:44:27 PM] Olga Caprotti: JS: to do a translation in the hybrid system (WP7) [11/9/12 1:45:37 PM] Olga Caprotti: JS: to ask deadlines to meritxell and email to maria [11/9/12 1:48:05 PM] Olga Caprotti: JvA: WP12 no exception to report, dissemination concerning innovation is now ready and sent, almost ready for distribution, progress on wp12 about training for new guys working on project, first week of dec. deliverable for May is no problem - Yuri will work on this. if meeting in dec. takes place, then no problem - we are on track. [11/9/12 1:48:12 PM] Olga Caprotti: LC is joining [11/9/12 1:49:28 PM] Olga Caprotti: LC: WP9 we have a coordinator for doing evaluation - both translation and usability. Jussi seems knowlegeable in the area, Jussi is his name. [11/9/12 1:50:48 PM] Olga Caprotti: LC: Jussi has requested the first evaluation corpus. [11/9/12 1:51:06 PM] Olga Caprotti: LC: we could run the script mentioned in the mail [11/9/12 1:51:48 PM] Olga Caprotti: MM: can Jussi contact us to let us know what to do? [11/9/12 1:52:29 PM] Olga Caprotti: NEF: Jussi must also contact us. [11/9/12 1:54:02 PM] Olga Caprotti: LC: the translation tool has new database in place, better timings. term tools by end of year [11/9/12 1:54:26 PM] Olga Caprotti: LC integration is going on in parallel. [11/9/12 1:55:02 PM] Olga Caprotti: LC: D 3.3 will be discussed next monday with inari, raw data already available [11/9/12 1:55:41 PM] Olga Caprotti: AR: UGOT - krasimir working on robust parsing - will be integrated in the architecture for cristina for the hybrid evaluation [11/9/12 1:55:53 PM] Olga Caprotti: AR: Thomas working on compiler [11/9/12 1:56:10 PM] Olga Caprotti: AR: new theoretical ideas have sped up the compilation 3 times [11/9/12 1:57:15 PM] Olga Caprotti: AR: JC has worked on the syntax editor - can change the translations and propagate in the tree - part of translator tools [11/9/12 1:57:46 PM] Olga Caprotti: AR: we can add new languages to the cultural heritage [11/9/12 1:58:01 PM] Olga Caprotti: MM: we need also the phrases for the web site [11/9/12 1:58:34 PM] Olga Caprotti: LC: MM plc contact jussi on this for crowdsourcing. [11/9/12 2:00:49 PM] Aarne Ranta: now I will leave for another meeting but please go on. And nice weekend! [11/9/12 2:01:22 PM] Mateva: nice weenkend for you too, Aarne! [11/9/12 2:01:49 PM] Olga Caprotti: thank you! [11/9/12 2:01:55 PM] Joris van Aart: thanks Olga! [11/9/12 2:02:26 PM] Olga Caprotti: JS: contract for aries is over - she is searching for a job. [11/9/12 2:03:14 PM] Olga Caprotti: FYI: OC's contract ends feb 28. [11/9/12 2:03:27 PM] Olga Caprotti: what happens afetrwards is still unclear [11/9/12 2:03:33 PM] Olga Caprotti: you might be on your own. [11/9/12 2:03:39 PM] Joris van Aart: have a nice weekend [11/9/12 2:03:44 PM] Olga Caprotti: bye [11/9/12 2:03:53 PM] Olga Caprotti: you too joris [11/9/12 2:03:58 PM] Jordi Saludes: Nice weekend to you all [11/9/12 2:04:10 PM] Olga Caprotti: bye all [11/9/12 2:04:25 PM] Lauri Carlson: Bye bye [11/9/12 2:04:30 PM] Jordi Saludes: Bye
Present:
Excused: Lauri
AGENDA
MINUTES
MXW is officially out of the Consortium, the EC has been notified and MOLTO needs to find a new partner. We approached IRF but the PO, upon inspection, has found this a risky choice. The EPO has been suggested by the PO but Aarne has also been previously contacted by them. He sent email and received very positive feedback. It is not sure if they will join as a partner. They are looking into the details of the workplan and will let us know ASAP. EPO would meet the EU requirement.
According to our workplan, we are behind schedule slightly so the workplan will need to be amended in the next report.
UPC needs the patent corpus for carrying out experiments - they will start with an artificial corpus. UGOT needs to develop more grammar tools before getting into the patent case.
Ontology interoperability is proceeding according to plan. Proof of concept showed that more effort than expected is needed to derive the grammars for covering the queries. Corpus of sentences used for evaluation will be collected and is based on a limited ontology - it will illustrate the NL queries that are expected. From this, we will produce guidelines for non experts entering the queries. Krasimir is visiting Sofia and will go to Ontotext to clarify usage of GF.
Kangaroo problems are planned as word problems. Question of hiring personel ahead of WP start. Change of the workplan requested.
Aarne leaves the meeting.
OC reminds of D10.1 - please collect planned activities.
Deliverables are all in dissemination.
Participants: LC, SN, JS, BP
Progress report is almost done - LC sent PMs for UHEL, OC will make PDF and deliver it.
AR: Ontotext has 12M in WP2, so how to use this. AR sent suggestion about using Eclipse plugin and give feedback, after writing GF grammars in connection with ontologies.
BP: connection to RDF DB we should do. About the Eclipse plugin, we will check what we should do there by middle next week.
AR: 12M seems a lot to be done in 4M (time left in WP2 according to workplan).
LC: what is this about using Eclipse. AR: John Camilleri has implemented it and released it. LC: we have Chinese people doing a Chinese GF grammar.
LC: UHEL has worked on translator's tools, we have a plan how this will work together, around Krasimir's editor. We got the editor working under Eclipse (?), we are adding the term editor to it. We would like to embed Krasimir's editor to GlobalSight. So now we have the MOLTO editor as one of the tools. We do not have yet a web demo but will come soon. The work on embedding the editor of Krasimir to other CMSs too as a PlugnPlay component. TODO: Mockup of MOLTO translation environment.
JS: Cristina's presented patent translation at MTXII - Demo available on ontotext website. Ares is doing the SAGE grammar to command CAS systems, almost done - interaction is a problem, servers fail fail after the 5th or 6th questions asked on the MAC, seems to work in Linux. Geogebra interface is planned but my contact is not answering. OC: will forward the CFP to Jordi, possibly fix a meeting with Dom and Jordi.
AR: Ontotext had good start with what Milen sent me.
BP: we have some momentum, museum wp is being handled by Mariana.
AR: UGOT released the Eclipse plugin as part of the gramamr IDE, integration of example-based grammar writing with the editor is under progress. GF3.3 released yesterday. Patents and museum workpackages - we sent off hybrid, modular architecture description to Cristina and Meritxell and are waiting for comments - we have a clear idea where we can improve SMT by GF. Nick has worked on Russian grammars. RBMT organization for June 2012. Will visit originators of workshop in Alicante next week. We start organization soon. Enlargement and extension of MOLTO by 3 months. Start date still under discussion.
LC, AR: UHEL an UZH cooperation has to be defined. Start work on multilingual wiki connecting it to the translation tools. LC looked into the Attempto tools. Technology transfer will be necessary as a stepping stone. Estonian work of Kaarel with Inaari will fit very well. We are discussing tele-working.
Any EACL papers on the way? Demo deadline is a month later. UGOT sent a tutorial proposal (will know next week).
Paper submitted to LIHMT-2011 on evaluation of hybrid systems.
WP4 Work done - two way grammar-ontology interoperability - public demo - http://molto.ontotext.com/ future work: - Experiment with automatic GF generation from ontology. Experiment with different ontologies.
WP7 Work done: - selected ontologies for semantic annotations - customized interface from wp4 for the beta prototype - http://molto-patents.ontotext.com/ add queries for patents future work: - Annotate patents and add them to the repository. - experiments with and refining selected ontologies. - beta prototype - Dec 2011. - Free text search functionality.
WP8 Work done: - Gothenburg museum data is processed, transformed, loaded and combined with dbpedia - http://museum.ontotext.com/ - experiments with Gothenburg museum data. future work: - use new painting ontology and new data from Gothenburg. - beta prototype - February 2012
Participants: OC, AR, JS, MC
Patent retrieval internal report was circulated today. MC is working on the prototype for d7.1, will be ready for tomorrow. Status is good, we will be ready for December as planned.
SMT group worked on D7.1 and a beta version is online for WP7. Cristina tried to translate some medical patents. EPO still not responding on the demands related to corpus, so they used American corpus from WT WP6, on D6.2, working on making assumptions, and coping variables, basically dealing with context. Run-time context used in sage computations. AR, Russian grammar almost complete for wp6, NF said it was much easier than the one for the Phrasebook.
MC in WP6 Ontotext has some months, maybe we need to meet to discuss - JS: interaction with ATP systems would be good. Discuss the kind of reasoning needed on the ontology to see if shallow reasoning provided by Ontotext engines is enough to deal with word problems to be tackled by WP6. MC, in WP4 we had email exchange about D4.3. We worked on WP7 and WP8, preparing a prototype for cultural heritage (Mariana is working with Dana). For WP2 we started testing the plugin, and are looking how to add visualizing tools for ontologies to the plugin. The meeting for WP3 needs LC feedback - will ask Seppo or Inari.
AR in the kickoff meeting we will focus on this new work in the GF eclipse plugin to show it to BI partner.
AR tutorial proposal at ICFP sent, turned down the one sent to EACL. UGOT has 2 papers submitted to EACL and a demo paper. 2 papers to LREC. Post proceedings of CNL will contain the paper by Kra and Ramona on Sumo Ontology and the Phrasebook paper.
OC to push publication more, check with BP whether Ontotext will exhibit MOLTO work in some conference.
Local organization of MOLTO-E Kickoff Meeting, in Gothenburg 12-13, 2011. Planned GF tutorial using Eclipse Plugin.
AR request to estimate the PMs used so far in each WP, to plan the figures we need to fill in if we request a workplan modification. For WP2 we like to extend it 3 months, move last deliverable.
Changes to the workplan
Lauri Carlson, Olga Caprotti, Milen Chechev, Aarne Ranta, Norbert E. Fuchs, Jordi Saludes, Jeroen van Grondelle.
[3/2/12 9:57:25 AM] Olga Caprotti: hello [3/2/12 9:57:37 AM] Olga Caprotti: milen, is this the right skype id? [3/2/12 9:57:54 AM] Jeroen van Grondelle: the right one for Jeroen ;-) [3/2/12 9:57:57 AM] Olga Caprotti: borislav, leave the conversation if you will not be able to be in the call [3/2/12 9:59:38 AM] Milen Chechev: this skype id is ok for the meeting, I'll change it after that :) [3/2/12 10:00:57 AM] Olga Caprotti: ok, everyone. [3/2/12 10:01:51 AM] Olga Caprotti: i would like to know who is on a fast connection, i am calling from home [3/2/12 10:02:00 AM] Olga Caprotti: eg Jordi are you at the uni? [3/2/12 10:02:08 AM] Jordi Saludes: yes, I am [3/2/12 10:02:25 AM] Olga Caprotti: aarne are you at the uni? [3/2/12 10:02:34 AM] Olga Caprotti: call should be hosted by fasted connection [3/2/12 10:02:46 AM] Olga Caprotti: fastest [3/2/12 10:02:59 AM] nefuchs: I am at home with a fast connection. [3/2/12 10:03:16 AM] Aarne Ranta: [3/2/12 10:07:44 AM] Olga Caprotti: Aarne: quite a few from UGOT, thomas demo of the grammar IDE, john's demo of the Eclipse plugin. [3/2/12 10:08:16 AM] Olga Caprotti: olga: lets bring old posters to the meeting [3/2/12 10:08:48 AM] Olga Caprotti: jordi: demo on wp6 [3/2/12 10:09:35 AM] Olga Caprotti: js to discuss demo of cristina and meritxell [3/2/12 10:10:06 AM] Olga Caprotti: milen: demo for wp7, incl. some work from upc -- do not know if ready [3/2/12 10:10:51 AM] Olga Caprotti: lauri: 5 things to demo if we can fix them before the meeting [3/2/12 10:11:03 AM] Olga Caprotti: global sight translation system [3/2/12 10:11:30 AM] Olga Caprotti: translation editor and equivalence editor (tabular editor conneted to ?forge) [3/2/12 10:11:44 AM] Olga Caprotti: ontology editor is most difficult one [3/2/12 10:11:58 AM] Olga Caprotti: seppo's ontology extraction system on its way [3/2/12 10:12:33 AM] Olga Caprotti: 5 people from UHEL [3/2/12 10:13:07 AM] Olga Caprotti: norbert: ace wiki with gf extension demo [3/2/12 10:15:38 AM] Olga Caprotti: pls create biblio [3/2/12 10:15:40 AM] Olga Caprotti: http://www.molto-project.eu/node/add/biblio [3/2/12 10:16:00 AM] Olga Caprotti: choose Miscellaneous [3/2/12 10:16:29 AM] Olga Caprotti: i will enter whatever comes my way [3/2/12 10:18:07 AM] Olga Caprotti: modify program to add demos to end of program [3/2/12 10:18:21 AM] Olga Caprotti: remove first demo session from coffee break [3/2/12 10:22:06 AM] Olga Caprotti: OC: color tag badges with tasks before registration [3/2/12 10:23:48 AM] Olga Caprotti: jeroen: bi demo not yet with gf integration [3/2/12 10:28:18 AM] Olga Caprotti: AR: workgroups, TH webserver tools, OC website tutorial [3/2/12 10:28:55 AM] Olga Caprotti: JS: acewiki work on wp6 [3/2/12 10:29:31 AM] Olga Caprotti: AR: best practices for grammar writing, because we need input from everyone, incl. BI [3/2/12 10:29:54 AM] Olga Caprotti: LC: ontology grammar interface discussion [3/2/12 10:30:16 AM] Olga Caprotti: LC: what happens when we edited some words? we need to figure out [3/2/12 10:31:06 AM] Olga Caprotti: MC: agree we need ontology-grammar interoperability group, had some conversation with Dana and Krasimir, need more discussion on verbalization [3/2/12 10:31:46 AM] Olga Caprotti: J : unfortunately not techn. represented but very interested [3/2/12 10:35:13 AM] Olga Caprotti: OC, AR: financial deps should do their work. [3/2/12 10:35:22 AM] Olga Caprotti: AR: who is coming to the review? [3/2/12 10:36:31 AM] Olga Caprotti: KK for UZH, BI not clear if possible to attend [3/2/12 10:37:37 AM] Olga Caprotti: AR: BI aspect is exploitation [3/2/12 10:38:32 AM] Olga Caprotti: Ontotext: Borislav and Milen [3/2/12 10:38:47 AM] Olga Caprotti: UPC: Jordi and Lluis [3/2/12 10:38:57 AM] Jordi Saludes: (fresh news from Cristina: no demo for wp5, for wp7, Milen will demo it) [3/2/12 10:39:00 AM] Olga Caprotti: UHEL: LC (again!) [3/2/12 10:39:15 AM] Olga Caprotti: UGOT: AR and OC [3/2/12 10:40:25 AM] Olga Caprotti: MC: should we provide an agenda? [3/2/12 10:41:15 AM] Olga Caprotti: AR: the PO will want the agenda, we will decide it during the business meeting [3/2/12 10:41:28 AM] Olga Caprotti: OC: circulate the review presentation structure [3/2/12 10:42:27 AM] Olga Caprotti: JG: we have a draft requirement document, online, use: biblio [3/2/12 10:42:41 AM] Olga Caprotti: biblio-deliverable [3/2/12 10:44:24 AM] Olga Caprotti: JG: we have an outline coming, like feedback [3/2/12 10:44:52 AM] Aarne Ranta: Call ended41 minutes 34 seconds
The meeting of April has been cancelled due to Easter vacation.
(skype chat log please edit parts missing due to bad quality)
[5/4/12 10:01:27 AM] Aarne Ranta: hello [5/4/12 10:01:47 AM] Aarne Ranta: [5/4/12 10:02:13 AM] Olga Caprotti: [5/4/12 10:02:56 AM] Olga Caprotti: Response to the review, action plan Project's flagships status Dissemination: upcoming events GF course in Apeldoorn FreeRBMT Exploitation venues Other matters [5/4/12 10:04:02 AM] Olga Caprotti: http://www.molto-project.eu/wiki/minutes-meetings/sg-meeting-20120504 [5/4/12 10:04:47 AM] Olga Caprotti added Seppo to this conversation [5/4/12 10:04:57 AM] Jeroen van Grondelle: I hav e troublw getting online [5/4/12 10:05:06 AM] Jeroen van Grondelle: I am now outside, next to a noisy fountain [5/4/12 10:05:23 AM] Jeroen van Grondelle: I will try to join now, but it might be a problem with the background noise [5/4/12 10:06:33 AM] Olga Caprotti: Status report from each node [5/4/12 10:07:20 AM] Olga Caprotti: AR: forthcoming EAMT and LREC presentation, busy organizing FreeRBMT, and Apeldoorn [5/4/12 10:07:58 AM] Olga Caprotti: AR: Esrebet is a new phd student, she is working on patent grammars, now concentrating on lexicons [5/4/12 10:08:26 AM] Olga Caprotti: AR: she works on German [5/4/12 10:09:41 AM] Olga Caprotti: AR: KA, working on robust translation and JC is working on AceWiki, SV is working on extracting lexicon from Wordnet [5/4/12 10:10:20 AM] Olga Caprotti: AR: we will use the same methodologyon WordNet from UHEL. We need to set a date. [5/4/12 10:10:41 AM] Aarne Ranta: Erzsébet [5/4/12 10:12:27 AM] Olga Caprotti: JS: Cristina is in touch with Pluto's team to discuss evaluation of Patent Evaluation, also preparing EAMT paper, I work on AceWiki and moved all tasks related to MGL and demo and debugging to Aries [5/4/12 10:13:37 AM] Olga Caprotti: JS: word problems formalization in GF will proceed in parallel with AceWiki while doing the integration. [5/4/12 10:14:02 AM] Olga Caprotti: AR: is the tutorial system integrated with the wiki [5/4/12 10:14:21 AM] Olga Caprotti: JS: not yet, I am trying to understand how AceWiki is working. [5/4/12 10:17:51 AM] Olga Caprotti: JS: idea is to leverage AceWiki to be able to add sentences, unmbiguously, enter them as assertion, check their consistency, (expect unique solution), student must be able to provide a different compatible model, and system will check the sentence based representation and the students based representation in terms of equations. The system will check equivalence of models. Plan to use CAS to check equational theories. [5/4/12 10:18:50 AM] Jordi Saludes: John Tinsley de Pluto [5/4/12 10:19:01 AM] Jordi Saludes: (is the one at Pluto contacted) [5/4/12 10:19:07 AM] Olga Caprotti: MC: attended WWW2012 conference with 2 presentations! [5/4/12 10:21:19 AM] Olga Caprotti: MC: got questions and answered all of them, Ontotext working on review's remarks to resubmit new versions of deliverables - UGOT and UHEL must be involved in new version of deliverables [5/4/12 10:21:49 AM] Olga Caprotti: MC: UGOT and UHEL have some extra resources to help with new versions [5/4/12 10:22:12 AM] Olga Caprotti: AR: pls send your requests to the site leaders of UGOT and UHEL [5/4/12 10:22:49 AM] Olga Caprotti: [5/4/12 10:23:26 AM] Olga Caprotti: AR: cc to Olga for getting disciplined :) she'll keep us on the line [5/4/12 10:23:59 AM] Olga Caprotti: MC: will write more about the tool generating the grammars [5/4/12 10:24:32 AM] Olga Caprotti: AR: must make sure this work in integrated with similar work in translator's tools and in semantic wiki [5/4/12 10:25:20 AM] Olga Caprotti: BP (previous were MC): released OWLIM 4 [5/4/12 10:27:22 AM] Olga Caprotti: JvG : paper for CNL about MONET-MOLTO sinergy, end of may review due [5/4/12 10:28:05 AM] Olga Caprotti: JvG: some of it will end in our next deliv. the requirement document. [5/4/12 10:29:13 AM] Olga Caprotti: JvG: public will have less details than consortium-only [5/4/12 10:29:48 AM] Olga Caprotti: AR: good to tell why BI is doing this to the public [5/4/12 10:30:14 AM] Olga Caprotti: JvG: paper writing went well but a bit time tight. [5/4/12 10:30:32 AM] Olga Caprotti: AR: CNL paper is relevant to grammar-ontology interoperability [5/4/12 10:31:07 AM] Olga Caprotti: AR: Milen you should see that paper [5/4/12 10:31:23 AM] Olga Caprotti: JvG: TODO: upload submitted version to website [5/4/12 10:32:12 AM] Olga Caprotti: JvG: we have a milestone, not started yet the work. We need to start working on this ASAP. I need to know contact point for the Java runtime [5/4/12 10:32:25 AM] Olga Caprotti: OC: we need Who's who in GF [5/4/12 10:32:59 AM] Olga Caprotti: AR: pls include me in the request for this milestone. Will figure out who is the right contact person [5/4/12 10:33:44 AM] Olga Caprotti: AR: fix dates for event in Apeldoorn [5/4/12 10:34:17 AM] Olga Caprotti: JvG: we wrap up the dates next week early [5/4/12 10:34:47 AM] Olga Caprotti: AR: we will be Ramona, John and Aarne - to cover all topics, maybe not overlapping all 3 [5/4/12 10:35:37 AM] Olga Caprotti: JvG: what we did for paper in GF grammars is beyond what I expected we could do. [5/4/12 10:37:47 AM] Olga Caprotti: NEF [5/4/12 10:38:14 AM] nefuchs: I dropped out [5/4/12 10:38:36 AM] Olga Caprotti: will you try again [5/4/12 10:39:09 AM] Olga Caprotti: i am on hold [5/4/12 10:39:11 AM] Jordi Saludes: Now I'm in hold [5/4/12 10:39:12 AM] Olga Caprotti: :) [5/4/12 10:40:13 AM] Olga Caprotti: i closed my conversation - aarne can you call again please? [5/4/12 10:43:43 AM] Jordi Saludes: Please call me again: I'm out. [5/4/12 10:44:18 AM] Olga Caprotti: when i try i get call on hold [5/4/12 10:44:32 AM] Olga Caprotti: aarne i think you need to close previous call [5/4/12 10:44:36 AM] Olga Caprotti: it is now on hold [5/4/12 10:45:32 AM] Aarne Ranta: Call ended43 minutes 41 seconds [5/4/12 9:58:00 AM] Olga Caprotti added Aarne Ranta, borislav popov, Jordi Saludes, nefuchs to this conversation [5/4/12 9:59:27 AM] Olga Caprotti added Lauri Carlson to this conversation [5/4/12 9:59:35 AM] Olga Caprotti added Jeroen van Grondelle to this conversation [5/4/12 9:59:53 AM] Olga Caprotti: http://www.molto-project.eu/wiki/minutes-meetings/sg-meeting-20120504 [5/4/12 10:00:04 AM] Olga Caprotti: Response to the review, action plan Project's flagships status Dissemination: upcoming events GF course in Apeldoorn FreeRBMT Exploitation venues Other matters [5/4/12 10:01:14 AM] Olga Caprotti: hello everyone [5/4/12 10:01:19 AM] Jordi Saludes: hi [5/4/12 10:01:24 AM] Olga Caprotti: I have no confirmation that Lauri will attend [5/4/12 10:01:40 AM] nefuchs: Hi [5/4/12 10:01:49 AM] Olga Caprotti: I see Aarne is on another [5/4/12 10:40:22 AM] Aarne Ranta: [5/4/12 10:40:41 AM] Aarne Ranta: restarted call [5/4/12 10:42:17 AM] Aarne Ranta: [5/4/12 10:42:47 AM] Aarne Ranta: [5/4/12 10:44:18 AM] Aarne Ranta: Olga, are you calling? [5/4/12 10:44:39 AM] Aarne Ranta: [5/4/12 10:44:48 AM] Aarne Ranta: [5/4/12 10:45:30 AM] Olga Caprotti: we are in a different call [5/4/12 10:46:21 AM] borislav popov added Milen Chechev to this conversation [5/4/12 10:47:07 AM] Olga Caprotti: NEF: we do not want CNL submission to be public [5/4/12 10:49:40 AM] Olga Caprotti: NEF: there could be a conflict now that a paper is withdrawn after it has been submitted [5/4/12 10:52:17 AM] Olga Caprotti: discussion on issue of withdrawl [5/4/12 10:52:24 AM] Olga Caprotti: OC: i cannot understand you [5/4/12 10:52:40 AM] Olga Caprotti: NEF: we have 14(?) papers [5/4/12 10:52:52 AM] Olga Caprotti: I cannot write minutes with this sound quality [5/4/12 10:53:09 AM] Jordi Saludes: (I cannot understand a thing) [5/4/12 10:53:22 AM] Aarne Ranta: I can't hear you Norbert [5/4/12 10:53:30 AM] Olga Caprotti: NEF we could not make out any word [5/4/12 10:54:02 AM] Olga Caprotti: NEF: guess 10 papers accepted [5/4/12 10:54:24 AM] Olga Caprotti: NEF: one invited speaker paid by MOLTO ...? [5/4/12 10:55:14 AM] Olga Caprotti: NEF: dealing with finding the invited speaker [5/4/12 10:55:46 AM] Olga Caprotti: NEF: Ace in GF, KK and JC worked well, about 90% subset of Ace [5/4/12 10:56:03 AM] Olga Caprotti: NEF: we should be ready by end of May [5/4/12 10:56:20 AM] Olga Caprotti: NEF: draft version of D11.1 [5/4/12 10:56:59 AM] Olga Caprotti: NEF: Kaarel working on demo version of ? [5/4/12 10:57:15 AM] Olga Caprotti: (please edit minute if you are able to) [5/4/12 10:57:50 AM] Olga Caprotti: NEF: I am overviewing the work [5/4/12 10:58:30 AM] Olga Caprotti: NEF: employment situation - contract will appear hopefully [5/4/12 10:58:52 AM] Olga Caprotti: NEF: will be part time so that total time will cover PMs budgeted for me [5/4/12 10:59:11 AM] Olga Caprotti: NEF: salary discussions ongoing [5/4/12 11:00:42 AM] Olga Caprotti: AR: the flagship was a reaction to the review - to provide nice well-rounded off demos [5/4/12 11:01:04 AM] Olga Caprotti: AR: demos will be independent of WP but show integration of the results [5/4/12 11:01:22 AM] Olga Caprotti: AR: we will have technical captain of the flagship [5/4/12 11:01:31 AM] Olga Caprotti: AR: natural one is AceWiki [5/4/12 11:01:50 AM] Olga Caprotti: AR: translators' tools is a big flagship [5/4/12 11:02:01 AM] Olga Caprotti: byw Borslav [5/4/12 11:02:36 AM] Olga Caprotti: NEF: I can use with JS on the reasoner's side [5/4/12 11:02:55 AM] Olga Caprotti: NEF: want to extend Race to handle the example of JS [5/4/12 11:03:18 AM] Olga Caprotti: NEF: will make interface between Race and AceWiki [5/4/12 11:03:46 AM] Olga Caprotti: AR: so will surprise the reviewers [5/4/12 11:04:58 AM] Olga Caprotti: http://www.molto-project.eu/wiki/living-deliverables/d101-dissemination-... [5/4/12 11:06:53 AM] Olga Caprotti: OC: we discuss exploitation during next meeting [5/4/12 11:07:10 AM] Aarne Ranta: Call ended22 minutes 20 seconds
Some notes about WP12 in telco: Lot of work on Paper for CNL 2012 We will now turn that into Spec, combined with outline I sent earlier. Due End of May Question: Project or Public? Answer: Public. We will need to present a public deliverable, with enough context for it to be useful also in exploitation. Details can be shared separatly, project confidential. Java integration: Due end of June - Discuss with Aarne who is liaison. Course: Dates should be fixed early next week.
[12/7/12 8:57:56 AM] Olga Caprotti added Aarne Ranta, Jeroen van Grondelle, Jordi Saludes, Lauri Carlson, Mateva, nefuchs to this conversation [12/7/12 8:58:13 AM] Olga Caprotti: Agenda ====== 1. status report 2. upcoming deliverables 3. review & final project meeting 4. dissemination activities 5. exploitation/future projects [12/7/12 8:59:08 AM] Olga Caprotti: Olga Caprotti set topic to "MOLTO SG November-Dec call" [12/7/12 9:00:15 AM] Olga Caprotti: hello - Jeroen migt be a few minutes late due to weather [12/7/12 9:00:22 AM] Jordi Saludes: ok [12/7/12 9:02:31 AM] Lauri Carlson: hello all [12/7/12 9:02:37 AM] Jordi Saludes: hello [12/7/12 9:03:11 AM] Olga Caprotti: hello Lauri, can you read the previous messages, starting with the Agena? [12/7/12 9:03:15 AM] Olga Caprotti: Agenda [12/7/12 9:03:25 AM] Lauri Carlson: Yes [12/7/12 9:03:28 AM] Olga Caprotti: ok [12/7/12 9:03:37 AM] Olga Caprotti: maybe we just wait until 10 past [12/7/12 9:04:39 AM] Mateva: good morning everyone :) [12/7/12 9:04:46 AM] Olga Caprotti: :) [12/7/12 9:04:51 AM] Jordi Saludes: good morning [12/7/12 9:05:03 AM] nefuchs: good morning [12/7/12 9:05:04 AM] Mateva: I just wanted to ask if we have some meetings per WP [12/7/12 9:05:22 AM] Mateva: I would be happy if we have a meeting on WP8 [12/7/12 9:05:38 AM] Mateva: very short one, dont want to waste anyone's time [12/7/12 9:05:56 AM] Mateva: but can we schedule one [12/7/12 9:06:06 AM] Olga Caprotti: Call started, 31 seconds [12/7/12 9:06:26 AM] Mateva: I know Dana will be available on 12th Dec [12/7/12 9:06:41 AM] Olga Caprotti: Call ended, 31 seconds [12/7/12 9:07:28 AM] Olga Caprotti: aarne, do you have the video call button - top right? [12/7/12 9:08:17 AM] Olga Caprotti: Olga Caprotti has ejected Aarne Ranta from this conversation [12/7/12 9:08:25 AM] Olga Caprotti added Aarne Ranta to this conversation [12/7/12 9:08:50 AM] Olga Caprotti: Aarne, do you see this now? [12/7/12 9:11:07 AM] Olga Caprotti: aarne? [12/7/12 9:13:02 AM] Olga Caprotti: Call started, 57 minutes 32 seconds [12/7/12 9:13:53 AM] Olga Caprotti: Agenda ====== 1. status report 2. upcoming deliverables 3. review & final project meeting 4. dissemination activities 5. exploitation/future projects [12/7/12 9:16:32 AM] Olga Caprotti added Joris van Aart to this conversation [12/7/12 9:17:01 AM] Olga Caprotti: WP8 request from Maria to Aarne. [12/7/12 9:18:04 AM] Olga Caprotti: AR: will have a meeting in Bielefeld with Jeoren from MOLTO and MONNET people, WP12 related and ontology related [12/7/12 9:18:41 AM] Olga Caprotti: LC: late deliverable D3.3. in the wirting now [12/7/12 9:19:19 AM] Joris van Aart: Joris Be Informed checked in [12/7/12 9:19:26 AM] Olga Caprotti: LC: evaluation - jussi has put together plan for methodology, contacts with many of you, corpus by end of year - he has not gotten everything yet [12/7/12 9:20:01 AM] Olga Caprotti: AR: how is the integration of translation tools in the global sight? [12/7/12 9:20:09 AM] Olga Caprotti: LC: problem of time mostly [12/7/12 9:21:04 AM] Olga Caprotti: AR: wrt translation tools, syntax editor is now added to the translation tools - looks quite useful to change thing sin the middle of the tree [12/7/12 9:21:33 AM] Olga Caprotti: AR: available in the minibar - integrated already in the minibar [12/7/12 9:21:58 AM] Olga Caprotti: LC: d3.3. is late [12/7/12 9:23:57 AM] Olga Caprotti: LC: target end of year for comment release of D3.3 [12/7/12 9:24:10 AM] Olga Caprotti: LC: tool will not be available in parallel [12/7/12 9:24:25 AM] Olga Caprotti: JS: late for WP6, expect publish for comments next week [12/7/12 9:25:38 AM] Olga Caprotti: JS: posted tickets to track but no answer [12/7/12 9:26:23 AM] Olga Caprotti: AR: i should have gotten these mails. LA has fixed the emailing [12/7/12 9:26:33 AM] Olga Caprotti: AR: will look into the tickets [12/7/12 9:27:03 AM] Olga Caprotti: LC: you may ff to IL if she could help solve some tickets [12/7/12 9:27:21 AM] Olga Caprotti: JS: not enough data for the evaltuation corpus in WP6 [12/7/12 9:27:43 AM] Olga Caprotti: JS: the word problem especially has too little lexicon [12/7/12 9:27:58 AM] Olga Caprotti: JS: logic of system will be the same but lexicon will increase [12/7/12 9:28:23 AM] Olga Caprotti: JS: ares has finshed contract - i cannot do German concrete [12/7/12 9:29:04 AM] Olga Caprotti: JS: if RGL works then adding German will be semi automatic [12/7/12 9:29:13 AM] Olga Caprotti: JS: work is mainly evaluation [12/7/12 9:29:44 AM] Olga Caprotti: JS: lexicon is small, 10s of words, maybe a bit involved [12/7/12 9:29:57 AM] Olga Caprotti: AR: pls make clear the task [12/7/12 9:31:00 AM] Olga Caprotti: OC: pls add some small documentation [12/7/12 9:31:24 AM] Olga Caprotti: AR: pls send me instructions so i can try to coordinate this work and fix the bugs [12/7/12 9:31:58 AM] Olga Caprotti: JS: first draft will contain documentation, currently only work on English [12/7/12 9:32:27 AM] Olga Caprotti: JS: adding ES and CA will be easy fo rme, i know [12/7/12 9:33:01 AM] Jordi Saludes: Dialog*.gf [12/7/12 9:33:12 AM] Olga Caprotti: JS: beginning of next week a first draft of del to give some documentation [12/7/12 9:33:15 AM] Jordi Saludes: mgl/wproblems [12/7/12 9:33:39 AM] Olga Caprotti: AR: we are not naming consistently the folders in SVN [12/7/12 9:34:10 AM] Olga Caprotti: JS: about WP7, working on translations in all directions, not any new work [12/7/12 9:34:41 AM] Olga Caprotti: JS: generation of patents in XML has some problems - the XML format is giving some issues [12/7/12 9:35:32 AM] Olga Caprotti: JS: detected a problem in annotation and translation of chem. compound is affecting workflow [12/7/12 9:35:51 AM] Olga Caprotti: JS: work with ontotext to detect chem. compounds [12/7/12 9:36:58 AM] Olga Caprotti: MM: work on WP7 issue with the format of the annotation and the initial markup of patents in XML -- we do the semantic annotation, they do the translation [12/7/12 9:37:50 AM] Olga Caprotti: MM: the pipeline for maintaining the original format of the patents - the online demo has new features [12/7/12 9:38:37 AM] Olga Caprotti: MM: full text search is now integrated in the demo, the source of doc is exposed, either original or translated [12/7/12 9:39:40 AM] Olga Caprotti: MM: will ask BP if he has shown any MOLTO work during events [12/7/12 9:40:22 AM] Olga Caprotti: MM: demo has improved from my pov [12/7/12 9:41:17 AM] Olga Caprotti: MM: integration with grammars in WP8 has been done, can write a question in NL, but our KB is too small in the grammar [12/7/12 9:42:01 AM] Olga Caprotti: MM: are we going to show only these descriptions in the grammar? [12/7/12 9:42:15 AM] Olga Caprotti: AR: we want more info from the museum and from europeana [12/7/12 9:42:54 AM] Jordi Saludes: (leave for 1 min) [12/7/12 9:43:02 AM] Olga Caprotti: JS: when Dana starts she will add things, and demo will improve [12/7/12 9:43:12 AM] Olga Caprotti: OC: http://www.wikipaintings.org/ [12/7/12 9:44:34 AM] Jordi Saludes: (again with you) [12/7/12 9:45:01 AM] Olga Caprotti: AR: for testing the query grammar and the ontology we can work with small KB [12/7/12 9:45:28 AM] Olga Caprotti: AR: we might want to add new predicates and functions [12/7/12 9:46:14 AM] Olga Caprotti: MM: wonder if enough description of the painting [12/7/12 9:46:34 AM] Olga Caprotti: MM: KB is very sparsely populated [12/7/12 9:47:25 AM] Olga Caprotti: MM: Mariana is working with Dana on a chapter of a book [12/7/12 9:48:39 AM] Olga Caprotti: JvA: Jeroen still stuck on snow [12/7/12 9:49:02 AM] Olga Caprotti: JvA: course in Aperldoorn, Kaarel will come on Tuesdya - elaborate program [12/7/12 9:49:15 AM] Olga Caprotti: JvA: efficient use of grammar resources on Tue and Wed [12/7/12 9:49:35 AM] Olga Caprotti: JvA: afterwards we attend Bielefeld, we expect boost for WP12 [12/7/12 9:49:53 AM] Olga Caprotti: JvA: should clarify demos, resources problems is sorted out [12/7/12 9:50:09 AM] Olga Caprotti: JvA: new resources will attend next week course [12/7/12 9:51:15 AM] Olga Caprotti: JvA: will travel to meetings [12/7/12 9:51:24 AM] Olga Caprotti: AR: was hoping to fix dates [12/7/12 9:51:56 AM] Olga Caprotti: LC: suggested week is ok - will enter doodle [12/7/12 9:52:06 AM] Olga Caprotti: AR: please add your avail. to doodle [12/7/12 9:52:14 AM] Olga Caprotti: MM: will add to doodle [12/7/12 9:52:34 AM] Olga Caprotti: Norbert? [12/7/12 9:54:06 AM] Olga Caprotti: AR: jordi - any preference on days of the week? [12/7/12 9:54:24 AM] Olga Caprotti: AR: norbert wants end of week [12/7/12 9:54:46 AM] Olga Caprotti: AR: 23-24 may suggested [12/7/12 9:55:24 AM] Olga Caprotti: AR: we ahve to send invitation to PO and scientific board. will contact PO with prel. date [12/7/12 9:56:43 AM] Olga Caprotti: AR: we will need a list of new projects we are involved in related to MOLTO [12/7/12 9:57:09 AM] Olga Caprotti: AR: UGOT takes part in 2propos. deadl 15 jan. [12/7/12 9:57:30 AM] Olga Caprotti: AR: but more calls are open, UPC and BI have things going on. [12/7/12 9:57:37 AM] Olga Caprotti: Hello norbert! [12/7/12 9:57:44 AM] Olga Caprotti: NEF: problems with mic. [12/7/12 9:58:09 AM] Olga Caprotti: NEF: progress - covering 15 languages, anaphora and ambiguity issues [12/7/12 9:58:20 AM] Olga Caprotti: NEF: show up during translations [12/7/12 9:58:41 AM] Olga Caprotti: NEF: D11.2 will be submitted mid jan because holidays [12/7/12 9:59:08 AM] Olga Caprotti: NEF: need comment round? [12/7/12 9:59:16 AM] Olga Caprotti: AR: yes - consortium agreement [12/7/12 9:59:27 AM] Olga Caprotti: NEF: then delive. will come ready end jan [12/7/12 9:59:48 AM] Olga Caprotti: NEF: evaluation - exchange of email, picture is getting clear on what we need to do [12/7/12 10:00:37 AM] Olga Caprotti: NEF: selected previously used examples by Tobias, chosen to show all features of grammar - but sentences not looking natural, maybe we change examples with museum [12/7/12 10:00:59 AM] Olga Caprotti: NEF: evaluation deli. planned for april, but is start is march [12/7/12 10:01:13 AM] Olga Caprotti: NEF: too late for us to integrate results of evalutation [12/7/12 10:01:25 AM] Olga Caprotti: AR: question to LC [12/7/12 10:01:58 AM] Olga Caprotti: AR: evaluation quality of some examples, we need them soon [12/7/12 10:02:15 AM] Olga Caprotti: LC: a couple of months from when we get the corpus [12/7/12 10:02:26 AM] Olga Caprotti: LC. could be end of feb if we get corpus now [12/7/12 10:02:37 AM] Olga Caprotti: AR: we could schedule priority [12/7/12 10:02:54 AM] Olga Caprotti: NEF: tobias can use january to setup eval. framework [12/7/12 10:03:06 AM] Olga Caprotti: NEF: we can start feb instead of mar [12/7/12 10:03:46 AM] Olga Caprotti: NEF: we need support as test persons for usability [12/7/12 10:04:14 AM] Olga Caprotti: AR: yes - everyone will test usability [12/7/12 10:04:38 AM] Olga Caprotti: NEF: by end of jan we will have input from tobias [12/7/12 10:06:01 AM] Olga Caprotti: NEF: tobias will be avail. 2.5 PM, we have 2.5PM left, 9 candidates for the position, we evaluate them next week, interviews next week, with UZH candidates -- [12/7/12 10:06:20 AM] Olga Caprotti: NEF. we have a choice - i think we will find someone [12/7/12 10:07:05 AM] Olga Caprotti: NEF: KK and I attended workshop in BZ, relation between CNL, authoring etc... talks on AceWIki-GF and race [12/7/12 10:07:12 AM] Jordi Saludes: btw, SX and JS submitted a paper to Tbilisi Mathematical Journal [12/7/12 10:07:30 AM] Olga Caprotti: NEF: up to 20 ppl attendance [12/7/12 10:07:30 AM] Jordi Saludes: on Multilingual Sage [12/7/12 10:08:06 AM] Olga Caprotti: NEF: will attach report to minutes [12/7/12 10:08:38 AM] Olga Caprotti: MM: need jussi contact [12/7/12 10:09:04 AM] Olga Caprotti: MM: i need list of tasks for us [12/7/12 10:09:19 AM] Olga Caprotti: MM: task and schedule [12/7/12 10:09:49 AM] Olga Caprotti: AR: next meeting will be in january [12/7/12 10:09:57 AM] Olga Caprotti: AR: merry Xmas and happy new year [12/7/12 10:10:36 AM] Olga Caprotti: Call ended57 minutes 32 seconds
[1/29/13 10:06:49 AM] Aarne Ranta: lauri: deliverable coming these days, evaluation in good speed [1/29/13 10:07:28 AM] Mateva: hello [1/29/13 10:08:22 AM] Mateva: Is the call scheduled for today? [1/29/13 10:12:01 AM] Aarne Ranta: lauri: City of Vantaa possible exploitation [1/29/13 10:13:36 AM] Aarne Ranta: jordi: wp6 finished on time, wp7 going on with German compounding etc [1/29/13 10:15:54 AM] Olga Caprotti: NEF: sent notes to OC - will copy to minutes [1/29/13 10:16:01 AM] Aarne Ranta: norbert: 11.2 open to discussion [1/29/13 10:16:23 AM] Norbert E. Fuchs: The public demo of AceWiki-GF is at http://attempto.ifi.uzh.ch/acewiki-gf/. [1/29/13 10:20:31 AM] Aarne Ranta: joris: BI visit to Bielefeld [1/29/13 10:27:07 AM] Olga Caprotti: MM: migrated the demos to (?), open source sw. Mariana working on a book chapter. We have to finish the patent demo and evaluate it - new version of query grammar not there yet. [1/29/13 10:29:19 AM] Olga Caprotti: MM: some queries return no answer. is this desirable or not? [1/29/13 10:32:56 AM] Olga Caprotti: MM: doing evaluation on museum case - [1/29/13 10:33:22 AM] Olga Caprotti: AR: for patents, was agreed Ontotext makes evaluation and exploitation analysis for it [1/29/13 10:33:53 AM] Olga Caprotti: AR: helsinki looks at translation quality rather than exploitation [1/29/13 10:34:43 AM] Olga Caprotti: MM: demo at molto.ontotext.com has been improved - comments welcome [1/29/13 10:35:13 AM] Olga Caprotti: MM: for answer grammar we have no translations [1/29/13 10:35:30 AM] Olga Caprotti: MM: can UGOT help? [1/29/13 10:35:44 AM] Olga Caprotti: AR: UGOT will help - but not urgent [1/29/13 10:37:07 AM] Olga Caprotti: AR: we worked on museum evaluation texts, and on patents (Ramona and Cristina) - results for En-De hybrid system evaluation are encouraging [1/29/13 10:37:44 AM] Olga Caprotti: AR: expected improvement with some new things - always taking some days to get results from evaluation [1/29/13 10:38:36 AM] Olga Caprotti: AR: we are doing robust translation with WordNet lexica, interesting for testing use of existing resources [1/29/13 10:39:20 AM] Olga Caprotti: AR: we will demo it [1/29/13 10:39:44 AM] Olga Caprotti: AR: looking into finnish wordnet, need to talk to you LC [1/29/13 10:40:16 AM] Olga Caprotti: AR: JC has integrated the syntax editor with online demo, minibar (javascript) will be good for translators tools [1/29/13 10:41:18 AM] Mateva: Olga, we migrated the demos to OWLIM 5 http://www.ontotext.com/taxonomy/term/40 (our semantic datastore) and Forest 1.4.1 (web interface for exploring it) [1/29/13 10:44:10 AM] Norbert E. Fuchs: UZH: Kaarel & Tobias submitted a paper on AceWiki-GF to ESWC 2013 [1/29/13 10:44:30 AM] Olga Caprotti: planned publications: UHEL, jussi-mari-lauri paper on national translation seminar / there is MT summit also in September [1/29/13 10:45:12 AM] Olga Caprotti: JS: Tblisi journal of mathematics submission required minor changes. [1/29/13 10:48:07 AM] Olga Caprotti: MOLTO is funded by the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement FP7-ICT-247914. [1/29/13 10:49:18 AM] Olga Caprotti: http://www.molto-project.eu/content/page/gf-application-grammars [1/29/13 10:51:49 AM] Olga Caprotti: BI will probably have a chapter in the MONNET book appearing in spring edited by SV. we will have 2 chapters there. [1/29/13 10:56:10 AM] Olga Caprotti: OC and AR: status of flagship [1/29/13 10:57:23 AM] Olga Caprotti: discussion on online seminar [1/29/13 10:59:28 AM] Olga Caprotti: let us schedule this
Mateva 10:02 AM good morning Olga Caprotti 10:05 AM Agenda status report email aarne AR: working with finishing deliverabe in WP8 with ontotext AR: and hybrid translation with UPC - AR: new PO is called Pierre-Paul Sondag Aarne Ranta 10:06 AM Pierre-Paul.Sondag@ec.europa.eu Olga Caprotti 10:08 AM AR: emails shared only for confidential matters - he is expected to email only with coordinator AR: would be good if some partners come to BCN one day ahead for last project meeting (wedn) LC: last friday had call, integration in progress - LA is working on it, IL workig on TF to GF translation LC: sticking to GlobalSight LC: other platforms are just computer translation tools and translators cannot change the source original, whereas I always thought MOLTO tools need to allow for feedback during the working time. Also it has to be a community and not a single translator, this is what GS does. The workflow is the most interesting part. LC: community aspect of usage of MOLTO tools is the interesting part LC: using grammars to populate translation memories is work in progress - LA started on that AR: pls change the date of D3.3 LC: D3.3. will be due by 31 March 2013 at the very latest LC: concern by Jussi on evaluation AR: Jussi should send some TODO MM: would like to have TODOs for each WP evaluation LC: checklist evaluation probably LC: bilateral discussions on how to carry out evaluations AR: pls use the google group to keep everyone informed JS: WP5 and WP6 no news to report, WP7 evaluation is being organized JS: concerned about flagship and wiki, we are supposed to use the wiki for the flagship - eg how to do the latex OC: we will use mathjax OC: linear algebra topic for demo MM: WP7 prototype being finalized. Ontotext requested many changes to the grammar to cover the ontology MM: retrieval system is being optimized (snippetting) MM: plan to finish draft of final deliverables by end of Feb, but delayed, if we need to do internal evaluation, then deadline will not be met. D7.3 is basically beingheld due to evaluation. MM: also working on WP8 so prefer to concentrate on WP7 first with short deadline AR: for the patent queries, RE made a SPARQL translation - does it work? MM: yes, system accepts both old and new queries. coverage of new grammar not so large - lexicon help needed AR: basically polishing details MM: WP8 had a deadline for query grammar to end of month. we worked on bulgarian mainly AR: query grammars all in place for WP8 - goal to have same languages as for the texts MM: fix what to adopt - old or new generation in SPARQL MM: prefer the new way to generate SPARQL directly MM: draft of KB for WP8 MM: mariana is working on MNL chapter for book AR: should wrapt up the deliverables and focus on demo NEF: summary sent OC: will be in minutes NEF: what sharing should be with Jussi and ourselves in evaluation NEF: UZH will do usability test and UHEL will do translation quality NEF: test sentences for Jussi ready - but grammatically ok but semantically nonsense NEF: long discussion on user evaluation of the wiki - we will use geography in ES DE EN NEF: vocabulary of 500 words, problem of proper names (google group discussion ongoing) NEF: we will have test persons writing sentences (choppy) NEF: deliverable 11.2 NEF: sw is permanently undr revision NEF: pointer to exact version of sw described in deliverable is already there as comment by Kaarel NEF: we are working on a constructive counter proposal to AR's flagship demos NEF: since many things in there have not been promised and we are unsure to be able to do them Lauri Carlson 10:45 AM I lost voice connection, reduced to reading the chat Sad Olga Caprotti 10:45 AM NEF: tobias decision to work half time allowed us to hire 2 extra persons. we are in a better shape with larger groups - lots of discussions JvA: evaluation results getting closer JvA: grammars on business models are final, models can be exported to OWL and back JvA: first explanation sentences can be generated - all in English for now - translations maybe within 1 or 2 weeks JvA: EN DUT ES DE targeted JvA: evaluation will be carried out on those sentences compared to our current approach JvA: flagship - we know what we want to show. Internal evaluation with BeInformed people - to see if explanations are better than before JvA: if you want to see our data for evaluation, let me know. JvA: will send to Jussi, EN and all langs when I get them JvA: performance evaluation and evaluate how to get grammars JvA: our current way to get explanations is very different from how we did in the past - how to evaluate it is new also. AR: are you specifying whcih skills are needed for each task? JvA: yes. we still need to finalize the integration in our tool in order to evaluate that. MM: is there a demo we can see? MM: the lemon integration is very interesting JvA: will post to the google group. JvA: setup a telco on exploitation MM: lets setup a meeting - Borislav must be present OC: volunteer for next google hangout 11:03 AM Call ended1 hour 1 minute
[3/27/13 2:57:04 PM] Olga Caprotti added Aarne Ranta, Jordi Saludes, Joris van Aart, Mateva, Norbert E. Fuchs to this conversation [3/27/13 2:57:54 PM] Olga Caprotti added Lauri Carlson to this conversation [3/27/13 2:57:58 PM] Olga Caprotti: hello [3/27/13 2:58:11 PM] Olga Caprotti: I added LAuri because I do not have Inari's contact [3/27/13 2:58:16 PM] Aarne Ranta: hello Olga [3/27/13 2:58:27 PM] Jordi Saludes: Hello [3/27/13 2:58:31 PM] Olga Caprotti: hello Aarne, I was preparing the call [3/27/13 2:58:49 PM] Olga Caprotti: Seppo is the deputy of Lauri [3/27/13 2:58:57 PM] Norbert E. Fuchs: Norbert: I cannot hear anything. [3/27/13 2:58:58 PM] Olga Caprotti added Seppo to this conversation [3/27/13 2:59:01 PM] Aarne Ranta: Inari is off line anyway [3/27/13 2:59:22 PM] Olga Caprotti: i can email to lauri and seppo [3/27/13 2:59:28 PM] Olga Caprotti: a reminder [3/27/13 2:59:38 PM] Mateva: hello, do you need inari's contact? [3/27/13 2:59:40 PM] Olga Caprotti: nef, we are not on the voice line [3/27/13 2:59:53 PM] Norbert E. Fuchs: OK [3/27/13 3:00:04 PM] Mateva added inarilistenmaa to this conversation [3/27/13 3:00:19 PM] Mateva: i will be back in 3 minutes [3/27/13 3:00:37 PM] Aarne Ranta: Olga will you start the call? [3/27/13 3:00:59 PM] Olga Caprotti: no aarne, i am on the home wifi, not high speed, it is best you host [3/27/13 3:01:08 PM] Aarne Ranta: OK I will [3/27/13 3:01:52 PM] Aarne Ranta: Call started, 5 minutes 36 seconds [3/27/13 3:04:26 PM] Mateva: is the call on? [3/27/13 3:05:36 PM] Olga Caprotti: yes maria - we are figuring it out [3/27/13 3:05:54 PM] Mateva: sorry, can you call me again [3/27/13 3:06:54 PM] Olga Caprotti: i just added you maria, do you see the call? [3/27/13 3:07:11 PM] Mateva: no [3/27/13 3:07:31 PM] Aarne Ranta: Call ended, 5 minutes 36 seconds [3/27/13 3:08:13 PM] Olga Caprotti: pls aarne can you call again [3/27/13 3:08:31 PM] Olga Caprotti: there was a misunderstanding with Maria [3/27/13 3:08:45 PM] Olga Caprotti: she and Joris are waiting fro the call [3/27/13 3:08:46 PM] Mateva: hm, I just tried to call in [3/27/13 3:08:55 PM] Olga Caprotti: you cannot do that on skype [3/27/13 3:09:10 PM] Jordi Saludes: I hung up too. [3/27/13 3:09:19 PM] Olga Caprotti: otherwise we all get another call [3/27/13 3:09:26 PM] Olga Caprotti: parallel to the one we are in [3/27/13 3:09:32 PM] Mateva: oh [3/27/13 3:09:37 PM] Olga Caprotti: Olga Caprotti has ejected inarilistenmaa from this conversation [3/27/13 3:09:59 PM] Olga Caprotti: aarne, we are only missing UHEL now [3/27/13 3:10:38 PM] Olga Caprotti: is anybody reading me here? [3/27/13 3:10:48 PM] Mateva: yes [3/27/13 3:10:48 PM] Norbert E. Fuchs: I do. [3/27/13 3:10:56 PM] Jordi Saludes: yes. But still no call for me. [3/27/13 3:10:57 PM] Aarne Ranta: Call started, 1 hour 11 minutes 57 seconds [3/27/13 3:11:19 PM] Olga Caprotti: do you see the topic of this conversation? [3/27/13 3:11:25 PM] Olga Caprotti: it says MARCH SG call [3/27/13 3:11:30 PM] Olga Caprotti: it is the current one [3/27/13 3:12:05 PM] Olga Caprotti: usual agenda [3/27/13 3:12:41 PM] Olga Caprotti: BI: (Joris) we have had 2 more explanation engines for ontologies [3/27/13 3:14:01 PM] Olga Caprotti: BI: we have sentences from old and from new prototypes - there are fragments produced by LEMON esp. lexicons, it is crucial for doing many languages however we can only do english so far- so we can only look at english - we had to come up with an alternative to do more langauges [3/27/13 3:14:18 PM] Olga Caprotti: BI: our data for evaluation is only in English - not so useful. [3/27/13 3:14:44 PM] Olga Caprotti: BI: we need to fix the LEMON lexication but this is a huge effort - quite a setback [3/27/13 3:15:45 PM] Mateva: Inari replied in mail [3/27/13 3:16:02 PM] Mateva: she is onlline now [3/27/13 3:16:12 PM] Olga Caprotti: BI: BI: we thought more languages would be supported by LEMON but when we tried to use it (in production) the only lexicon we could use was English. We need tagging - we might do it manually - for the evaluation for Dutch - our main lang pair was NL-EN [3/27/13 3:17:01 PM] Mateva added Inari Listenmaa to this conversation [3/27/13 3:17:08 PM] Olga Caprotti: AR: will you solve this before the review? [3/27/13 3:17:13 PM] Aarne Ranta added Inari Listenmaa to this conversation [3/27/13 3:17:17 PM] Olga Caprotti: BI: yes - we might have to do it manually [3/27/13 3:18:05 PM] Inari Listenmaa: Ok, I'm here [3/27/13 3:18:05 PM] Olga Caprotti: Inari did you just join? [3/27/13 3:18:10 PM] Inari Listenmaa: Mic problems probably once again [3/27/13 3:18:14 PM] Olga Caprotti: do you have a mic? [3/27/13 3:18:15 PM] Inari Listenmaa: But i can listen [3/27/13 3:18:20 PM] Inari Listenmaa: Yes [3/27/13 3:18:29 PM] Olga Caprotti: Olga Caprotti has ejected Lauri Carlson from this conversation [3/27/13 3:19:44 PM] Olga Caprotti: NEF: 3 isssues, evaluation : wiki gf logging and stability improved, design of evaluation was long discussion, i will be one of the external evaluator and there will beother 2 [3/27/13 3:19:59 PM] Olga Caprotti: NEF: we sent sentences to Jussi [3/27/13 3:20:15 PM] Olga Caprotti: NEF: we are waiting for feedback - [3/27/13 3:20:24 PM] Inari Listenmaa: Jussi will be traveling next week [3/27/13 3:20:30 PM] Inari Listenmaa: But he's got them [3/27/13 3:20:36 PM] Olga Caprotti: NEF: acewiki gf is using a GF service on our service [3/27/13 3:21:08 PM] Olga Caprotti: NEF: flagship, please play around and send us feedback [3/27/13 3:21:26 PM] Olga Caprotti: NEF: sent report by email - (will be in the minutes) [3/27/13 3:24:00 PM] Olga Caprotti: MM: WP4 ended long time ago but we are working an annex on a deliverable due on 5th april. we have other two more urgent deliverables on may 31st - the new doc will clarify details of the grammar-ontology interoperability: we expect the part on generation sparql gf from UGOT [3/27/13 3:25:13 PM] Olga Caprotti: MM: we would like the partners' contribution to extend wbk to ita, fre, ger. in WP4 -- verbalization of predicates in the ontology. [3/27/13 3:26:19 PM] Olga Caprotti: MM: aarne pls take a look at how we do the mapping from the ontology predicates to gf types [3/27/13 3:29:50 PM] Olga Caprotti: MM: eclipse plugin -- we promised some bug fixing, early April. Not really using it because of the many issues [3/27/13 3:30:06 PM] Olga Caprotti: MM: lexicons needed for the grammar for the demo [3/27/13 3:30:26 PM] Olga Caprotti: MM: examples in the demo need the lexicons absolutely [3/27/13 3:30:55 PM] Olga Caprotti: MM: cannot meet the evaluation deadline unless lexicons are in- please prioritize this. [3/27/13 3:32:17 PM] Olga Caprotti: MM: WP8 we are working on the prototype - lots of effort with Dana [3/27/13 3:32:35 PM] Olga Caprotti: MM: in my opinion it is looking great [3/27/13 3:32:52 PM] Olga Caprotti: MM: elegant generation of sparql - publication possible [3/27/13 3:33:20 PM] Olga Caprotti: MM: deadline for book extended to apr. 21st [3/27/13 3:33:53 PM] Olga Caprotti: oops [3/27/13 3:33:58 PM] Olga Caprotti: pls add me back [3/27/13 3:35:32 PM] Mateva: foo [3/27/13 3:35:36 PM] Aarne Ranta: yes [3/27/13 3:35:37 PM] Inari Listenmaa: I see the minutes also [3/27/13 3:35:40 PM] Jordi Saludes: me too [3/27/13 3:35:51 PM] Inari Listenmaa: Do you want me to add Jussi to the conversation? He has a mic that works [3/27/13 3:36:07 PM] Inari Listenmaa added Jussi Rautio to this conversation [3/27/13 3:37:25 PM] Olga Caprotti: AR: hello Jussi, do you have questions to ontotext+ [3/27/13 3:37:37 PM] Olga Caprotti: JR: no, have to think a bit [3/27/13 3:39:28 PM] Olga Caprotti: JS: cristina is finalizing the library for wp5, and the deliverable is work in progress - wp7 by meritxell, they are creating syntactic measures for the lexicons, olga seb and i working on the flagship for mathematics: we can express definition of a prime number. oc working on spanish and italian. we present it in madrid. [3/27/13 3:42:48 PM] Jussi Rautio: are there any guesstimates about when the maths flagship has some word problems to quality evaluate? [3/27/13 3:44:21 PM] Olga Caprotti: JS: d6.3 has to be made final - OC is reviewing it [3/27/13 3:44:40 PM] Inari Listenmaa: Btw when is the final molto meeting? [3/27/13 3:44:57 PM] Olga Caprotti: AR: do the final logistics fo rthe meeting [3/27/13 3:45:18 PM] Olga Caprotti: OC: to do - finalize details of the MOLTO meeting [3/27/13 3:45:37 PM] Jordi Saludes: have to go out for 1 min. [3/27/13 3:45:49 PM] Olga Caprotti: @Inari: http://www.molto-project.eu/meeting/final-project-meeting [3/27/13 3:46:01 PM] Olga Caprotti: MM: we planned wedn thu fri [3/27/13 3:46:05 PM] Inari Listenmaa: Olga: Ok, thanks! [3/27/13 3:46:25 PM] Olga Caprotti: AR: we add 22nd for preparation [3/27/13 3:47:36 PM] Inari Listenmaa: Yeah [3/27/13 3:47:42 PM] Inari Listenmaa: Jussi can talk about 9 at least [3/27/13 3:47:51 PM] Olga Caprotti: AR: TODO apr. 3rd schedule a call for organizing the meeting (12:00-18:00) [3/27/13 3:48:21 PM] Olga Caprotti: JR: are we implementing the TAUS API for the hybrid systems? [3/27/13 3:48:38 PM] Olga Caprotti: AR: is is worth doing? [3/27/13 3:48:48 PM] Jordi Saludes: I'm back [3/27/13 3:48:59 PM] Olga Caprotti: JR: i think it would be easy to implement afterwards [3/27/13 3:49:13 PM] Olga Caprotti: JR: we have no resources in UHEL to implement it. [3/27/13 3:49:33 PM] Inari Listenmaa: LA is basically our only programmer and he's working 16.5 hrs/week [3/27/13 3:50:41 PM] Inari Listenmaa: I'd like to ask about the GF cloud things [3/27/13 3:51:22 PM] Inari Listenmaa: Aarne mentioned some time ago that UGOT is planning an integration of Syntax editor and Simple translation tool. Is this still happening? [3/27/13 3:51:48 PM] Aarne Ranta: This integration has higher priority than the TAUS standard [3/27/13 3:51:56 PM] Inari Listenmaa: I agree [3/27/13 3:53:04 PM] Inari Listenmaa: D3.3 deadline is sunday, can we write cloud toosl integration in it? [3/27/13 3:53:34 PM] Olga Caprotti: JR: looks like TAUS API is made more for project management than for translation [3/27/13 3:53:36 PM] Inari Listenmaa: We abandoned GS [3/27/13 3:53:43 PM] Inari Listenmaa: We're doing Pootle instead [3/27/13 3:53:47 PM] Olga Caprotti: ooooh [3/27/13 3:53:54 PM] Olga Caprotti: when do we get a demo? [3/27/13 3:54:08 PM] Inari Listenmaa: When LauriA has it ready [3/27/13 3:54:21 PM] Inari Listenmaa: He's the bottleneck, the only good programmer with 2 working days a week [3/27/13 3:54:32 PM] Inari Listenmaa: :D [3/27/13 3:54:33 PM] Inari Listenmaa: Alright [3/27/13 3:54:35 PM] Olga Caprotti: are you more satisfied with it? [3/27/13 3:54:48 PM] Inari Listenmaa: :---D [3/27/13 3:54:50 PM] Inari Listenmaa: No it's not [3/27/13 3:55:17 PM] Olga Caprotti: AR: general advice, make sure the due deliverables are clear, and honest [3/27/13 3:55:22 PM] Inari Listenmaa: Yes [3/27/13 3:56:05 PM] Inari Listenmaa: I'm sorry, I didn't realize nobody hadn't informed you [3/27/13 3:56:50 PM] Olga Caprotti: OC: i guess Pootle is good news [3/27/13 3:57:09 PM] Olga Caprotti: AR: Jussi are you still waiting for some eval material? [3/27/13 3:57:22 PM] Inari Listenmaa: @Olga: It is, already the first prototypes are much more simple and usable than GS ever was [3/27/13 3:57:38 PM] Olga Caprotti: JR: got patent material today - will look at it next week - not avail. before [3/27/13 3:58:15 PM] Olga Caprotti: JR: the most difficult thing is to find evaluators, they have to be professional [3/27/13 3:58:37 PM] Olga Caprotti: AR: how urgent are the ontotext samples? [3/27/13 3:58:56 PM] Olga Caprotti: JR: everything is now urgent - we are running out of time [3/27/13 3:59:35 PM] Olga Caprotti: JR: please send samples asap [3/27/13 4:00:15 PM] Olga Caprotti: AR: UGOT will look at the grammars, ontotext will prepare the texts [3/27/13 4:01:22 PM] Olga Caprotti: JR: about the query languages - i understand the museum query languages are sent to professionals - will they evaluate the linguistic quality? [3/27/13 4:01:47 PM] Olga Caprotti: AR: we will not have as many experts as we planned. will be able to do only for some languages [3/27/13 4:02:09 PM] Olga Caprotti: JR: can you create some queries for which the can get answers? [3/27/13 4:02:35 PM] Olga Caprotti: MM: i will send you the demo link. [3/27/13 4:02:42 PM] Mateva: museum.ontotext.com [3/27/13 4:03:11 PM] Olga Caprotti: MM: working wrt to dbpedia paintings but more will come [3/27/13 4:03:55 PM] Olga Caprotti: MM: it does not generate all kinds of sentences because some data is not available in dbpedia (eg color of painting) [3/27/13 4:04:46 PM] Olga Caprotti: MM: i am a fan of this grammar generating sparql [3/27/13 4:05:18 PM] Olga Caprotti: JR: math word problems - [3/27/13 4:05:53 PM] Olga Caprotti: JS: check the wiki http://www.molto-project.eu/wiki/living-deliverables/d63-assistant-solvi... [3/27/13 4:05:56 PM] Jordi Saludes: lot of echo for me [3/27/13 4:06:33 PM] Aarne Ranta: WP6 provide examples as a multilingual treebank to evaluate [3/27/13 4:07:29 PM] Olga Caprotti: @JR, MM: I am the contact point for requests for evaluation [3/27/13 4:07:54 PM] Olga Caprotti: AR: practice: send to the whole molto - he who feels responsible answers [3/27/13 4:09:32 PM] Inari Listenmaa: A second [3/27/13 4:09:51 PM] Inari Listenmaa: You can include me in the molto organizatory emails [3/27/13 4:09:57 PM] Inari Listenmaa: Seppo is having a baby in a few days [3/27/13 4:10:06 PM] Inari Listenmaa: Yes :P [3/27/13 4:10:27 PM] Olga Caprotti: ok Inari - you are now the deputy of Lauri then [3/27/13 4:10:47 PM] Inari Listenmaa: Yeah, if I had known of this meeting earlier, I would've brought my laptop [3/27/13 4:10:55 PM] Olga Caprotti: AR: DD presented the paintings today, also worked on hybrid systems [3/27/13 4:11:04 PM] Inari Listenmaa: By the way, did Maria mention that we've also produced input for D4.3A [3/27/13 4:11:19 PM] Olga Caprotti: AR: WP5 and WP3 overlap in the robust parser [3/27/13 4:11:57 PM] Olga Caprotti: AR: we'd like to send materials for eval using the robust parser [3/27/13 4:12:48 PM] Olga Caprotti: (I must go -- pls add minutes yourself here) I will cut&paste when back [3/27/13 4:12:55 PM] Olga Caprotti: bye - happy easter [3/27/13 4:12:57 PM] Inari Listenmaa: I can add minutes [3/27/13 4:13:21 PM] Inari Listenmaa: Bye! [3/27/13 4:15:25 PM] Inari Listenmaa: AR: using GF needs some training and learning [3/27/13 4:15:57 PM] Inari Listenmaa: AR: we want to quantify how much effort needed to train people to use GF [3/27/13 4:16:16 PM] Inari Listenmaa: AR: BeInformed has been aware of this, categorization of different level of skills [3/27/13 4:17:06 PM] Inari Listenmaa: MM: Borislav & Maria ( & Petar, milen) have some comments [3/27/13 4:17:34 PM] Inari Listenmaa: AR: we have found lots of little errors in rgl [3/27/13 4:17:45 PM] Inari Listenmaa: AR: main thing keeping me busy with gf [3/27/13 4:18:32 PM] Inari Listenmaa: AR: still TODO, some of the bugs found in ACE are RGL bugs [3/27/13 4:18:50 PM] Inari Listenmaa: JR: Phrasebook is also revealing RGL bugs [3/27/13 4:19:14 PM] Inari Listenmaa: JR: Polish Phrasebook problems; rgl problems, some things you can't even say in POlish [3/27/13 4:20:50 PM] Inari Listenmaa: MM: also in Bulgarian some phrases sounds strange. Hesitation in evaluating, tradeoff between "can be said like this, perfectly said like this". Important to advise people how to evaluate. [3/27/13 4:21:00 PM] Inari Listenmaa: JR: evaluators commented that there is no context [3/27/13 4:21:29 PM] Inari Listenmaa: I don't have anything to add either [3/27/13 4:22:51 PM] Inari Listenmaa: Bye [3/27/13 4:22:54 PM] Aarne Ranta: Call ended1 hour 11 minutes 57 seconds
[4/30/13 1:55:41 PM] Olga Caprotti added Aarne Ranta to this conversation [4/30/13 1:55:57 PM] Olga Caprotti added Jordi Saludes to this conversation [4/30/13 1:56:23 PM] Olga Caprotti added Mateva to this conversation [4/30/13 1:56:32 PM] Olga Caprotti added borislav popov to this conversation [4/30/13 1:56:41 PM] Olga Caprotti added Lauri Carlson to this conversation [4/30/13 1:56:51 PM] Olga Caprotti added Norbert E. Fuchs to this conversation [4/30/13 1:57:00 PM] Olga Caprotti: aarne [4/30/13 1:57:40 PM] Olga Caprotti: HERE [4/30/13 1:57:48 PM] Olga Caprotti: do you see this conversation? [4/30/13 1:58:03 PM] Aarne Ranta: Yes SG April Call [4/30/13 1:58:18 PM] Olga Caprotti: who else needs to be added? [4/30/13 1:58:23 PM] Aarne Ranta: Be Informed [4/30/13 1:58:32 PM] Olga Caprotti added Jeroen van Grondelle to this conversation [4/30/13 1:58:42 PM] Olga Caprotti: is lluis attending? [4/30/13 1:58:43 PM] Aarne Ranta: Or Joris? [4/30/13 1:58:50 PM] Aarne Ranta: Yes, Lluis too [4/30/13 1:59:07 PM] Jordi Saludes: Lluis want to attend: Please add him [4/30/13 1:59:22 PM] Olga Caprotti: i do not have his skype id [4/30/13 1:59:27 PM] Olga Caprotti: can you try to add him? [4/30/13 1:59:36 PM] Olga Caprotti: Joris is out of office [4/30/13 1:59:52 PM] Aarne Ranta added lluismarquez to this conversation [4/30/13 2:00:07 PM] Aarne Ranta: Shall I start the call [4/30/13 2:00:27 PM] Olga Caprotti: yes please [4/30/13 2:00:34 PM] Jordi Saludes: lluismarquez [4/30/13 2:00:49 PM] Aarne Ranta: Call started, 1 hour 40 minutes 37 seconds [4/30/13 2:01:16 PM] Olga Caprotti: very windy now [4/30/13 2:01:22 PM] Jordi Saludes: yes [4/30/13 2:01:33 PM] Olga Caprotti: I hear only static [4/30/13 2:01:39 PM] Jordi Saludes: better [4/30/13 2:01:45 PM] Olga Caprotti: yes, better now [4/30/13 2:02:06 PM] Norbert E. Fuchs: Static again. [4/30/13 2:02:24 PM] Mateva: yes [4/30/13 2:02:29 PM] Olga Caprotti: I will minute [4/30/13 2:02:29 PM] Norbert E. Fuchs: yes [4/30/13 2:03:13 PM] Olga Caprotti: Agenda [4/30/13 2:03:29 PM] Olga Caprotti: 1. status report: problems with last deliverables [4/30/13 2:03:42 PM] Olga Caprotti: 2. final project meeting schedule open day [4/30/13 2:03:49 PM] Olga Caprotti: 3. review organization [4/30/13 2:03:55 PM] Olga Caprotti: 4. other matters [4/30/13 2:04:24 PM] Lauri Carlson: can you hear me (Lauri) [4/30/13 2:04:28 PM] Olga Caprotti: AR: UGOT status report - we delivered d8.3 the cult heritage [4/30/13 2:04:46 PM] Olga Caprotti: AR: we worked on presentations, exceeded travel budget [4/30/13 2:05:05 PM] Olga Caprotti: AR: only 2 or 3 will attend final proj. meeting [4/30/13 2:05:28 PM] Lauri Carlson: can you hear me? [4/30/13 2:05:35 PM] Jordi Saludes: no [4/30/13 2:05:48 PM] Lauri Carlson: i can hear you but mic not working [4/30/13 2:06:02 PM] Olga Caprotti: UPC: from LM [4/30/13 2:06:42 PM] Olga Caprotti: LM: wp5 from cristina, worked mostly on deliverable, done at 99%, final polishing, basically ready - re-evaluation of the systems [4/30/13 2:07:09 PM] Olga Caprotti: LM: meritxell has worked on deliverable for wp7 - and done a paper for MT summit [4/30/13 2:07:59 PM] Olga Caprotti: LM: for wp5-wp9 evaluation done automatically but still need to do the human evaluation in wp9 [4/30/13 2:08:11 PM] Olga Caprotti: LM: this eval could be critical, has to be done manually [4/30/13 2:08:38 PM] Olga Caprotti: LM: for the flagship, would like a prototype demoing the translation of patents - we have a online demo [4/30/13 2:09:43 PM] Olga Caprotti: LM: we have a online translation engine, but it is slow - works on our machine, we have issues with local installation of GF, must port it to machine in UGOT. if we need to do it using the webservice, it is more work and we must discuss division of labor [4/30/13 2:09:51 PM] Aarne Ranta: patent server who when where [4/30/13 2:10:24 PM] Olga Caprotti: AR: ok for the patent server, who? when? where? important questions. [4/30/13 2:10:53 PM] Olga Caprotti: AR: manual eval of patents, difficult to find experts - they are looking for people [4/30/13 2:10:55 PM] Lauri Carlson: I changed to another machine, can you join me again [4/30/13 2:12:05 PM] Olga Caprotti: AR: @ontotext, do you have contacts for human evaluation of patents? [4/30/13 2:12:39 PM] Olga Caprotti: MM: might have resources for German, but not sure, must ask. Let us set some timeframe for this task [4/30/13 2:13:15 PM] Olga Caprotti: are we back? [4/30/13 2:13:43 PM] Lauri Carlson: in the chat yes, but no voice [4/30/13 2:14:51 PM] Olga Caprotti: lauri, do you hear us? [4/30/13 2:15:20 PM] Lauri Carlson: yes now I can but what about you hearing me [4/30/13 2:15:40 PM] Olga Caprotti: @LC, we will test when your report comes [4/30/13 2:16:01 PM] Aarne Ranta: Eng-Ger Eng-Fre [4/30/13 2:16:33 PM] Olga Caprotti: AR: we need human evaluator for patents translations in eng-ger, eng-fre [4/30/13 2:16:50 PM] Olga Caprotti: AR: task managed by Jussi, as part of WP9 [4/30/13 2:16:55 PM] Olga Caprotti: MM: will check [4/30/13 2:17:18 PM] Olga Caprotti: AR: will ask Jussi to send to you the scope of the task [4/30/13 2:17:38 PM] Olga Caprotti: AR: matter of hours [4/30/13 2:18:56 PM] Olga Caprotti: LM: regarding wp7, the system can do queries in eng and fre, no german yet [4/30/13 2:19:28 PM] Olga Caprotti: AR: also the french grammar is a bit flawed, ramona promised to fix it [4/30/13 2:19:38 PM] Olga Caprotti: AR: in the query grammars [4/30/13 2:20:31 PM] Olga Caprotti: JS: about wp6, dissemination to madrid, emath meeting. talked to open universities an dvirtual uni [4/30/13 2:20:47 PM] Olga Caprotti: JS: polishing installation of d6.3 [4/30/13 2:21:08 PM] Olga Caprotti: JS: trying ubuntu [4/30/13 2:22:12 PM] Olga Caprotti: OC: concerned about quality of italian - fixing it in mathtalk [4/30/13 2:22:24 PM] Olga Caprotti: AR: request of sentences [4/30/13 2:22:38 PM] Olga Caprotti: JS, OC: we have split the task [4/30/13 2:23:03 PM] Olga Caprotti: JS: we may include conjectures [4/30/13 2:23:50 PM] Olga Caprotti: AR: false statements could be mistaken as translation errors [4/30/13 2:24:05 PM] Olga Caprotti: NEF: proofs that fail reveal a lot of the theorem prover [4/30/13 2:26:02 PM] Olga Caprotti: JS: maybe we mark false statements by F to make sure they are not taken as translation mistakes [4/30/13 2:27:04 PM] Olga Caprotti: OC: presentation also under progress [4/30/13 2:27:13 PM] Olga Caprotti: JS: how many slides? [4/30/13 2:27:29 PM] Olga Caprotti: AR: more a limit on length of presentation [4/30/13 2:28:06 PM] Olga Caprotti: NEF: 6 minutes at the senate - just saw it today. [4/30/13 2:29:30 PM] Olga Caprotti: OC: google hangout call had no answers [4/30/13 2:29:44 PM] Olga Caprotti: MM: was waiting for such an invite [4/30/13 2:30:02 PM] Olga Caprotti: @MM, it is in the MOLTO page on +Google [4/30/13 2:30:25 PM] Olga Caprotti: AR: good to rehearse live in your group [4/30/13 2:30:31 PM] Mateva: I guess I get no notifications there... [4/30/13 2:30:52 PM] Olga Caprotti: OC: will invite MM to +Google page. [4/30/13 2:31:10 PM] Olga Caprotti: MM: working on wp7 and wp8. [4/30/13 2:31:44 PM] Olga Caprotti: MM: deliverables, 7.3 my work, and 8.3 mostly mariana and partly me. MT paper with txell [4/30/13 2:32:11 PM] Olga Caprotti: MM: museum prototype non finished, improve sparql grammar, clean dbpedia data, expose pictures [4/30/13 2:32:37 PM] Olga Caprotti: MM: we have national vacation 1-6 may [4/30/13 2:32:51 PM] Olga Caprotti: MM: would like to have painting images today [4/30/13 2:33:08 PM] Olga Caprotti: MM: for patents, still have to add german, will go fast when we have grammars [4/30/13 2:33:21 PM] Olga Caprotti: MM: d43a, deliverable [4/30/13 2:33:30 PM] Olga Caprotti: MM: some bug fixes to eclipse [4/30/13 2:33:52 PM] Olga Caprotti: OC: have not yet been able to use eclipse plugin [4/30/13 2:34:47 PM] Olga Caprotti: OC: have you used the Eclipse plugin? [4/30/13 2:35:00 PM] Olga Caprotti: MM: we created a wizard for it. [4/30/13 2:35:23 PM] Olga Caprotti: MM: our wizard is not generic enough, it is crafted for our SPARQL endpoint [4/30/13 2:36:09 PM] Olga Caprotti: MM: extra work will be delayed until demos are done [4/30/13 2:36:15 PM] Olga Caprotti: MM: screencast planned [4/30/13 2:39:00 PM] Olga Caprotti: MM: shall we have the pdf on the main page? [4/30/13 2:40:19 PM] Olga Caprotti: MM: about the voting? [4/30/13 2:40:41 PM] Olga Caprotti: AR: voting should be finished very soon [4/30/13 2:41:11 PM] Olga Caprotti: AR: we finish the voting today [4/30/13 2:41:30 PM] Olga Caprotti: OC: we do it by finalizing the program of the meeting [4/30/13 2:42:01 PM] Olga Caprotti: AR: my opinion is that all presentations are good enough [4/30/13 2:42:06 PM] Olga Caprotti: NEF: I agree [4/30/13 2:42:17 PM] Olga Caprotti: AR: many improvements still needed [4/30/13 2:42:56 PM] Olga Caprotti: NEF: also during the Barcelona meeting we will improve the presentations [4/30/13 2:43:20 PM] Olga Caprotti: AR: advisory board will comment too [4/30/13 2:43:57 PM] Olga Caprotti: MM: should we present the software or the screencast? [4/30/13 2:44:17 PM] Olga Caprotti: AR: let's have the films ready - more convincing if we show the sw [4/30/13 2:47:52 PM] Jordi Saludes: http://upcommons.upc.edu/ [4/30/13 2:47:55 PM] Mateva: http://www.ontotext.com/publications [4/30/13 2:48:10 PM] Norbert E. Fuchs: Olga, there is an echo of your voice. [4/30/13 2:48:16 PM] Olga Caprotti: OC: mention the open access archive [4/30/13 2:48:29 PM] Olga Caprotti: OC: reference list [4/30/13 2:49:45 PM] Olga Caprotti: NEF: d11.3 last one, we are working on it [4/30/13 2:50:13 PM] Olga Caprotti: NEF: expected 20 may - we have 30 participants in evaluation [4/30/13 2:50:25 PM] Olga Caprotti: NEF: some participants have not even done stage 1 [4/30/13 2:50:34 PM] Olga Caprotti: NEF: we wait until we get these [4/30/13 2:50:56 PM] Olga Caprotti: AR: there cannot be furtehr delay [4/30/13 2:51:06 PM] Olga Caprotti: NEF: tobias employment ends today [4/30/13 2:51:18 PM] Olga Caprotti: NEF: he continues part-time as supervisor [4/30/13 2:51:36 PM] Olga Caprotti: NEF: we sent sentences to Jussi but he was not happy with sentences that made not sense [4/30/13 2:52:28 PM] Olga Caprotti: NEF: still have to do screencast, KK reviewed all presentations, not so much feedback on our acewiki [4/30/13 2:52:57 PM] Olga Caprotti: NEF: we worked a bit on ace and ape, added exponentiation [4/30/13 2:53:13 PM] Olga Caprotti: NEF: KK, Victor, NEF will be in BCN [4/30/13 2:53:22 PM] Olga Caprotti: NEF: KK and NEF attend review [4/30/13 2:53:41 PM] Olga Caprotti: NEF: employment extend to mid june [4/30/13 2:53:58 PM] Olga Caprotti: NEF: @AR pls fix date of review [4/30/13 2:54:16 PM] Olga Caprotti: AR: i agree but the PO has not answered [4/30/13 2:54:33 PM] Olga Caprotti: AR: will ask again [4/30/13 2:55:28 PM] Olga Caprotti: AR: we will decide what to present at the review in BCN [4/30/13 2:56:20 PM] Olga Caprotti: LC: UHEL, wp9 jussi is hard pressed [4/30/13 2:56:36 PM] Olga Caprotti: LC: some work he cannot do because no material and no testers [4/30/13 2:56:47 PM] Olga Caprotti: LC: wp6 examples missing [4/30/13 2:57:11 PM] Olga Caprotti: LC: dutch materials from BI was never seen - probably not included [4/30/13 2:57:38 PM] Olga Caprotti: LC: patent case, no response from EPO as human evaluator of patents [4/30/13 2:57:49 PM] Olga Caprotti: LC: probably have to hire experts [4/30/13 2:59:11 PM] Olga Caprotti: LC: how about matrixware people? would they know patents translators? [4/30/13 2:59:21 PM] Olga Caprotti: OC: check with Seppo about Neil [4/30/13 3:00:14 PM] Olga Caprotti: LC: harvesting terms and converting to GF demo in the works [4/30/13 3:00:42 PM] Olga Caprotti: LC: inari has finished d3.3 - she has a version [4/30/13 3:00:57 PM] Olga Caprotti: OC: n.tipper@matrixware.com [4/30/13 3:01:44 PM] Olga Caprotti: LC: the platform chosen for it is already on our site for some years [4/30/13 3:02:16 PM] Olga Caprotti: OC: have some tools been integrated? [4/30/13 3:02:38 PM] Olga Caprotti: LC: the translations we produced were not interesting enough [4/30/13 3:03:12 PM] Olga Caprotti: LC: the robust translations could be a candidate [4/30/13 3:04:14 PM] Olga Caprotti: LC: the localization industry could probably be convinced more than the translation industry [4/30/13 3:05:10 PM] Olga Caprotti: LC: because of the Finnish regulation - we have to let people take holiday [4/30/13 3:05:24 PM] Olga Caprotti: LC: most people will be on holidays [4/30/13 3:07:19 PM] Olga Caprotti: http://www.molto-project.eu/event/molto-final-project-meeting-barcelona-... [4/30/13 3:21:13 PM] Lluís Màrquez: I have to leave. Still haven't had lunch and starving! We (Cristina-Meritxell-I) will respond to the final proposal of the schedule. Bye [4/30/13 3:21:25 PM] Olga Caprotti: ciao - thanks [4/30/13 3:22:28 PM] Mateva: sorry, I am on wireless in this room and it seems to be inconsistent [4/30/13 3:41:29 PM] Aarne Ranta: Call ended1 hour 40 minutes 37 seconds
Utrecht 21.9.2012
Work group day.
Inari will make a list of questions for all grammar writers to answer, and write about the results in D9.2, the final evaluation report.
D9.2 is due March 2013, let's set a deadline in February 2013.
D2.3 Best Practices
All MOLTO WPs that have grammars:
Ongoing projects: Monnet/Be Informed, Query(?),
Already completed grammars: Phrasebook, Museum, ACE(?), Patents(?), MGL(?)
Things to report:
Things to report:
Abstract syntax design
Concrete syntax implementation
This applies for all grammars, in progress or complete. Phrasebook was the first MOLTO grammar, in 2010 spring. We could compare it to newer grammars. MGL has been rewritten since WebALT; some reflections on that maybe. How about the evolution of Ontotext query grammar? First it was a bunch of rules to transform SPARQL-GF, only later transformed into a grammar where English, Swedish and SPARQL are all concrete syntaxes.
If some of the terms are completely unfamiliar to you, it doesn't matter. Just answer to those you have used.
Answer yes to all tools that you have used.
a) Text editor + GF shell
if yes, some more questions:
b) GF-Eclipse plug-in
c) Cloud-based IDE (http://cloud.grammaticalframework.org/gfse/)
Answer yes to all tools that you have used.
2.1 compilation diagnostics:
a) verbose (import -v FILE)
b) type check dumps (gf --dump-tc FILES)
c) Other, what?
2.2 grammar display modes:
a) print_grammar (pg) with different options. Examples below.
pg -words
Outputs all terminal tokens in the grammar. Example:
Hammurabi> pg -words , amahhas amtahas awiilam awiilim awiilum ašaqqal ašebbir aštebir ...... pay shall silver slave slave's slaves slaves' the
pg -missing
Shows what abstract syntax constructions are missing from which concrete syntaxes. Example:
Phrasebook> pg -missing PhrasebookCat : MMust ObjPlur ThesPlur ThesePlur ThosePlur PhrasebookEng : ObjPlur ThesPlur ThesePlur ThosePlur
b) abstract_info (ai)
ai EXP
Shows info about an expression. Example:
Phrasebook> ai HowFarFromBy fun HowFarFromBy : Place -> Place -> ByTransport -> Question ; Probability: 6.666666666666667e-2
c) show_opers (so)
Shows available opers for a type. Requires first import -retain FILE
. Example:
Hammurabi> i -retain HammurabiEng.gf Hammurabi> so NP defnt : CN -> NP indef : CN -> NP xOf : N2 -> CN -> NP xOfPron : N2 -> Pron -> NP
d) compute_concrete (cc)
Computes the result of an oper. Requires first import -retain FILE
. Example:
> import -retain ParadigmsGer.gf > compute_concrete -list mkN "Buch" "Bücher" neuter Buch, Buch, Buch / Buche, Buchs / Buches, Bücher, Bücher, Büchern, Bücher, ResGer.Neutr
a) treebanks
b) random generation
Answer yes to all tools that you have used.
a) RGL synopsis document http://www.grammaticalframework.org/lib/doc/synopsis.html
b) RGL source browser http://www.grammaticalframework.org/lib/doc/browse/
c) download the source files and read them
d) Library browser grammar to construct RGL expressions (LibraryBrowser.pgf here http://cloud.grammaticalframework.org/minibar/minibar.html)
These questions are more "how?" than yes/no. Short answers are enough.
5.1 Starting point for your grammar
a) Test corpus of sentences you want to express in the grammar
b) Abstract syntax based on an existing formal system
c) Other method of abstract syntax design
5.2 Abstract syntax
5.2.1 Basic unit of the grammar
a) Sentence
b) Paragraph or more
5.2.2 Semantic control
a) No semantic control in abstract syntax (John likes beer and beer likes John equal)
b) Simple, restrictive abstract syntax (e.g. HowFarBy : Place -> ByTransport -> Question
, so how far is John by beer? not allowed.)
c) Dependent types
5.3 Module structure
a) Base module and domain-specific extensions
A way to structure abstract syntax. Different abstract syntax files for base grammar and domain extensions, concrete syntax irrelevant for this question.
b) Functors
A way to structure concrete syntax. It can be used together with base grammar and domain extensions, or with only a single abstract syntax file.
5.4 Concrete syntax
a) lincats only from recommended RGL API categories, that is Text, Utt, S, QS, Cl, CN, NP, AP, Adv, Card
(or with good reason also VP, VPSlash
)
b) lincats from non-recommended categorires: lexical (N, A, V
) or structural (Det, Quant, RP
).
c) lincats from categories that you have created yourself (see pg. 31-32 in Best practices document)
For example, you need a record type such as oper NPNationality : Type = {lang : NP ; country : NP ; prop : A}
for all variants that are needed to express someone's nationality.
It is recommended to have user-friendly constructor in a base grammar, such as oper mkAuthor : Str -> Author = \a -> mkNP (mkPN a)
, and in domain extension use only mkAuthor "Rembrandt"
instead of mkNP (mkPN "Rembrandt")
.
d) Only API-level functions (ParadigmsXxx, SyntaxXxx) -- no tables, no concatenation, no MorphoXxx opers.
Participants: AR, JS, LC, MM, KK, OC, JvG
AR proposes that the last meeting will be in connection with the final review, towards the end of the project. Suggested place is Barcelona, need to be verified with the PO.
It has to be before the end of the project in order to be able to fund traveling to attend the meeting and review. We want to have all flagships represented. The schedule will include a 1st Public Day and a 2nd Discussion day and reflection of project. Then we will have the review - maybe after one day rest.
JvG is concerned on time management during review so that we do better than last time.
OC: to circulate the template of the reviewer form so that everyone can check how to best present the result for reviewers to be able to fill in efficiently the review form.
JvG also concerned about uniform look for presentation (MOLTO branding).
OC to prepare template for slitex or similar. JvG (and everyone else) to check again the last review's comments.
Proposed date for meeting: May 20-24, 2013
Follows a discussion on attending ESWeb (ESWC2013) as sponsor, whether it is in the budget. The cost now for sponsorship of such events is 3000 Eur. It was not planned in our MOLTO budget but possibly Ontotext has budgeted some money for exhibitions at events such as these. MM to discuss this with Borislav.
Request to everyone to have an overall view on the use of resources at their site. How many PMs are left and whether to shift efforts among workpackages.
UHEL reports having had problems filling in planned positions so they have money left for hiring. They will need to hire an admin person for translations but still they could have left resources for 1 extra position. They also need to solve some issues with new hiring policy at UHEL.
UZH reports that their person months are fine, Tobias will join in the coming period.
JvG reports that BI has used 25% of the budget, possibly this is less than planned.
Second week of March at UGOT: 9-11/3 (check with PO and wrt reviews). The same format as Varna. But more structure and focus in WP meetings: prepare a template, circulate questions in advance. Invite scientific advisors and PO.
Feedback to Lauri before 17 September on D9.1. And to Olga on D1.2.
MXW has left, submitted financial report, paid back remaining advances to UGOT, in May 2010. Reason: company out of business. PO was informed, SC agreed in Skype meeting. Remains formality of EC approving that MXW left. Negotiations with EPO going on, decision taken there in end October. Some data privately released to Cristina from IRF.
WP7 will be delayed to start M9. Maybe don't need to change end month, just compress.
WP6 start moved from M7 to M4. Jordi will decide about the other dates. Probably unchanged.
ICFP tutorial? EACL tutorial?
Ontotext will start show demo on their booths soon.
GF summer school 2011 in Barcelona? Not just resources but also apps. Everyone agrees this can be a MOLTO event.
Olga checks introduction of Drupal Listhandler. And provide FAQ for using Drupal.
This is a Skype meeting about WP11 (semantic wiki) on 13 December 2011 at 9am.
Kaarel, John, Krasimir, Tobias.
Tobias:
Kaarel:
John:
JPGF or GF web service?
Reasoning?
Grammar?
Adding new Words
More ideas:
Tobias:
Kaarel:
Participants: OC, MA, RE, CE, MG
Topic: how to exploit the current MOLTO results and tools in the general panorama of e-science and WDML, and CLARIN and .....
Minutes
Malin and Ramona have worked on developing a Swedish parser based on a wide coverage grammar, Ramona and Txell are handling the patents language, and Jordi the mathematical fragments, plus we have the KRI from Ontotext. It seems that all this should come together now :)
So how hard is the following challenge:
In discussing the above we stumbled on a number of questions:
A number of task have been volunteered :) here below. No specific deadline, maybe one of funding.
TODO log:
[4/4/12 11:24:51 AM] Olga Caprotti: http://eudml.org/home.action;jsessionid=4E3743F7FAF048E7CDA31E9ACD53E054 [4/4/12 11:39:35 AM] Olga Caprotti: OC: look for swedish math paper to send to malin [4/4/12 11:39:46 AM] Olga Caprotti: OC: look for funding [4/4/12 11:40:19 AM] Cristina: CE: look for lexicon catalan-spanish-english [4/4/12 11:41:11 AM] Olga Caprotti: RE: scenario of example based generation, check if existing swedish domain-specific language guide for instructors at uni [4/4/12 11:41:25 AM] Cristina: MG: look for the terminology unit at upc [4/4/12 11:41:59 AM] Olga Caprotti: OC: scenario for exploitation planning of MOLTO result: e-science multilingual data harvest [4/4/12 11:42:54 AM] Olga Caprotti: [4/4/12 11:43:32 AM] Olga Caprotti: MA: ask at språkbanken how specific domain terms make it into the dictionary
We will try to integrate the one-click system with the translation tools or a hybrid based on robust parsing if available. We can decide the final architecture after the deliverables. Thomas should help on this. Alternatively, we can set up an online demo that translates a (small) bunch of patents or raw text.
We have a complete version of the grammars that covers 57 different queries. Maria will give, by Wednesday, detailed feedback and instructions about any required update (either the language or the SPARQL)
Meritxell will create a makefile to easily compile the document. UGOT will write a description of the last version of the grammar. Ontotext will try to find meaningful examples and update the interface accordingly Aarne suggested to stress the different MT approaches as an executive summary: - hybrids systems are SMT -driven - query-language is grammar and ontology driven Also, include Erzsébet in the authors list
Cristina will complete the part related to hybridisation Krasimir is rewriting the robust parsing part Ramona will complete the part related to GF (Section 4) Shafqat will probably finish the experiments on robust parsing this week. The deliverable will be extended later with the analysis relatedto this experiments.
MT summit, demo track. All WP7 people involved. Since MT is the focus of the conference, we will present a general description of the prototype stressing the two approaches followed in the hybrid vs. the queries.
Journal paper on the hybrid: Cristina and all
Multi-word Units in Machine Translation and Translation Technology: Conference paper about lexicon? (Ramona et al )
MT Summit/Workshop on patents with the most relevant aspect seen in D5.3
Agenda
Participants: Aarne Ranta (UGOT), Adam Slaski (UGOT), Mariana Damova (Ontotext), Yasen Kiprov (Ontotext), Milen Chechev (Ontotext), Lluis Màrquez (UPC), Cristina España (UPC), Meritxell Gonzàlez (UPC).
The patents retrieval system requires an ontology on the particular domain. Retrieval will be based on the knowledge representation of the claims. MD informs that Ontotext has an ontology on the Biomedical domain but lacks an ontology on the particular domain for the WP7. UPC will provide Ontotext patent examples from the current corpus (Domain IPC A61P: Specific therapeutic activity of chemical compounds or medicinal preparations). Ontotext has an ontology capturing the structure of Patent documents. It consists of several modules including some FDA terms, drugs and
measurement related models.
The GF processes the patent claims and produces a representation in the GF abstract syntax. The GF can translate chunks and align the texts. These tasks will be used to build the hybrid MT system. Patents can be translated on a batch process, similarly to the retrieval pre-processing of the documents, but in a separate process.
Online (on the fly) translation occurs somehow in the query. The user writes a query in any of the available languages (theoretically English, French and German). This query will be translated into the GF abstract syntax. Then, the SPARQL query will be build from the abstract representation provided by the GF.
AR informs about the negotiation with EPO regarding the terms of use of the patents corpus (namely, the non‐commercial use of the applications). We will work for having a demo hosted by the EPO.
Attachment | Size |
---|---|
Minutes-WP7-11-June-2011.pdf | 65.54 KB |
The goal of the meeting is to discuss how to proceed with wp8 before the last deliverable d8.3.
Agenda
The data to be covered in d8.3 • additions of the lexicon entries • transform data to RDF triples • multilingual transformation
Harmonize the painting queries demo and the ontotext demo
• coverage of the query and answer grammar
Book chapter • we are suppose to submit 20 pages by the 31st of March
Evaluations and tracking • time lines
Participants: AR, DD, MM, OC, MD
[1/16/13 1:58:23 PM] Dana Dannélls added Aarne Ranta to this conversation [1/16/13 1:58:29 PM] Dana Dannélls added Mariana Damova to this conversation [1/16/13 1:58:32 PM] Dana Dannélls added Mateva to this conversation [1/16/13 1:58:37 PM] Dana Dannélls: Call started [1/16/13 1:59:16 PM] Olga Caprotti: we are 1 minute early [1/16/13 2:00:09 PM] Mariana Damova: Hi: am am going to join in a bit later ... [1/16/13 2:00:24 PM] Dana Dannélls: ok [1/16/13 2:04:19 PM] Olga Caprotti: DD: added wikipedia data txt to svn, ontotext will need to add this data to the KRI, is it ok? [1/16/13 2:04:55 PM] Mariana Damova: No, we have another better suggestion, we carry a service [1/16/13 2:05:05 PM] Mariana Damova: http://factforge.net [1/16/13 2:05:17 PM] Mariana Damova: http://www.ontotext.com/factforge [1/16/13 2:05:27 PM] Olga Caprotti: MM: we will have to make RDF triples for the new data [1/16/13 2:06:04 PM] Mariana Damova: the FactForge has already triples of DBpedia, which is the RDF version of Wikipedia and 9 more datasets [1/16/13 2:06:22 PM] Mariana Damova: we also have methods to link repositories [1/16/13 2:06:32 PM] Mariana Damova: I was about to send you SPARQL queries [1/16/13 2:06:52 PM] Mariana Damova: whihc are showing what kind of information they can carry [1/16/13 2:06:59 PM] Mariana Damova: they can bring [1/16/13 2:07:01 PM] Mariana Damova: back [1/16/13 2:08:46 PM] Olga Caprotti: OC: does DBpedia have same as wikipedia content? if so, you have RDF already [1/16/13 2:09:12 PM] Olga Caprotti: MM: I'd like demo working by end of january, including data specific samples [1/16/13 2:09:33 PM] Olga Caprotti: AR: we can also then work on optimizing the grammars for that demo [1/16/13 2:10:04 PM] Olga Caprotti: AR: we can translate museum names for this data set [1/16/13 2:10:48 PM] Olga Caprotti: AR. entity strings are now not translated in the grammar [1/16/13 2:11:17 PM] Olga Caprotti: AR, one would like to translate "City Museum" [1/16/13 2:11:39 PM] Olga Caprotti: DD: i offer to translate museum names, painters [1/16/13 2:11:51 PM] Olga Caprotti: AR: translate painting names is hard generally [1/16/13 2:12:28 PM] Olga Caprotti: AR: or transliterated [1/16/13 2:13:04 PM] Olga Caprotti: AR if we have such translations in Wikipedia we can use them [1/16/13 2:13:31 PM] Olga Caprotti: AR: MonaLisa becomes Gioconda in Italian [1/16/13 2:14:01 PM] Olga Caprotti: DD welcome Mariana [1/16/13 2:14:57 PM] Olga Caprotti: DD: tags for languages, a tag for eg PabloPicasso with Hebrew transliteration [1/16/13 2:15:33 PM] Olga Caprotti: DD for the entities we are translating on the RDF on the side [1/16/13 2:16:35 PM] Olga Caprotti: MM: so we take the RDF tags for the language we are using - hope this does not collide with Java [1/16/13 2:17:34 PM] Olga Caprotti: Agenda The data to be covered in d8.3 • additions of the lexicon entries • transform data to RDF triples • multilingual transformation Harmonize the painting queries demo and the ontotext demo • coverage of the query and answer grammar Book chapter • we are suppose to submit 20 pages by the 31st of March Evaluations and tracking • time lines [1/16/13 2:18:57 PM] Olga Caprotti: MD: what kind of data to use beside the goteborg data museum - from other data sets we have triples on artwork not on painting [1/16/13 2:20:09 PM] Mariana Damova: # Cities where paintings of Modigliani are located (PROTON) PREFIX dbpedia: PREFIX ff: PREFIX ptop: PREFIX pext: SELECT DISTINCT ?painting_l ?owner_l ?city_l WHERE { dbpedia:Amedeo_Modigliani pext:isAuthorOf ?painting . ?painting ptop:isOwnedBy ?owner ; ff:preferredLabel ?painting_l. ?owner ff:preferredLabel ?owner_l . ?owner ptop:locatedIn [ a pext:City ; ff:preferredLabel ?city_l ] } [1/16/13 2:20:23 PM] Mariana Damova: here is the query that you can try on http://factforge.net [1/16/13 2:20:25 PM] Olga Caprotti: AR: we must get the grammars done - especially the text generation -- for the evaluation task [1/16/13 2:21:17 PM] Olga Caprotti: AR: who will host the demo? [1/16/13 2:21:30 PM] Olga Caprotti: MD: ontotext if the demo is based on RDF [1/16/13 2:21:46 PM] Olga Caprotti: AR: when did you contact the museum ppl last? [1/16/13 2:21:57 PM] Olga Caprotti: DD: invited to my defence [1/16/13 2:22:33 PM] Olga Caprotti: AR: lets say we have a system that demos architecture and fuctionality for the deliverable, and make sure we expand the data for the flagship [1/16/13 2:24:40 PM] Olga Caprotti: MM: we can make russian and polish grammar, do not know about final quality - what is desirable [1/16/13 2:24:52 PM] Olga Caprotti: AR: full quality is what we want [1/16/13 2:25:20 PM] Olga Caprotti: DD: maybe a native speaker is not necessary to write the grammar. [1/16/13 2:25:31 PM] Olga Caprotti: AR: a native is necessary to evaluate the quality [1/16/13 2:26:04 PM] Olga Caprotti: MM: i was asked to evaluate a bulgarian grammar and i was very happy with it [1/16/13 2:26:48 PM] Aarne Ranta: check coverage of text data set [1/16/13 2:28:22 PM] Olga Caprotti: MD: which infrastructure will you use fo rthe data set? the relational kb or the rdf? [1/16/13 2:28:44 PM] Olga Caprotti: AR: at present to work on grammars we use relational but plan to use RDF in the final demo [1/16/13 2:29:10 PM] Olga Caprotti: MD: including the wikipedia data? [1/16/13 2:29:36 PM] Olga Caprotti: AR: we have some random instances that have to cover all possible cases of the text generation [1/16/13 2:29:49 PM] Olga Caprotti: AR: we will move to more realistic data afterwards [1/16/13 2:30:21 PM] Olga Caprotti: MM: can you send the data to me pls? [1/16/13 2:30:38 PM] Olga Caprotti: DD: i am working on that, will send it in a couple of days [1/16/13 2:32:17 PM] Olga Caprotti: DD: MD, you said we have 2 demos, we need only one. how are we going to harmonize the paintings on the ontotext demo? [1/16/13 2:32:45 PM] Olga Caprotti: AR: we have to implement a SPARQL on the ontotext demo [1/16/13 2:33:14 PM] Olga Caprotti: MD: we need time to invent the model on the transformation to SPARQL [1/16/13 2:33:24 PM] Olga Caprotti: MD: and time to implement it. [1/16/13 2:34:07 PM] Aarne Ranta: after January: finish SPARQL backend of UGOT query language [1/16/13 2:34:30 PM] Aarne Ranta: as the main method of harmonization [1/16/13 2:34:44 PM] Aarne Ranta: who is going to work with this? Maria? [1/16/13 2:35:13 PM] Mariana Damova: I am sorry, i had to hang up, because there was a lot of side noise ... [1/16/13 2:35:29 PM] Olga Caprotti: it is good now that you are gone [1/16/13 2:35:59 PM] Olga Caprotti: it might have been a doppler effect from your loudspeakers [1/16/13 2:36:51 PM] Olga Caprotti: MM: not deployed under the demo yet but we have been working on mapping from NL to SPARQL [1/16/13 2:37:24 PM] Olga Caprotti: MM: we need to generate lots of resources to get this working - created some scripts to do it automatically [1/16/13 2:37:38 PM] Olga Caprotti: MM: now blocks me the lack of RDF to experiment with [1/16/13 2:38:06 PM] Olga Caprotti: AR: can't you generate RDF from the samples currently in the grammar [1/16/13 2:38:37 PM] Olga Caprotti: AR: all the data is in the relational db [1/16/13 2:38:37 PM] Mariana Damova: back again [1/16/13 2:38:53 PM] Olga Caprotti: AR: enough to experiment with the semantics [1/16/13 2:39:36 PM] Olga Caprotti: DD: that example you sent me was based on a record from the goteborg city museum, was good [1/16/13 2:40:26 PM] Olga Caprotti: MM: descriptions are too sparse in current records, my NL is not generated if parts of the records are missing [1/16/13 2:41:10 PM] Olga Caprotti: DD: i added a file called gcm.haskell with more information about the fields that are required [1/16/13 2:41:28 PM] Olga Caprotti: DD: i also sent translations in english of all the fields [1/16/13 2:42:15 PM] Dana Dannélls: GCMDataPainting.hs [1/16/13 2:42:38 PM] Olga Caprotti: MM: will revise the data then - i have not used the new files - i will use these [1/16/13 2:47:11 PM] Olga Caprotti: AR: whether we would have obligatory painter and painting is given [1/16/13 2:48:04 PM] Olga Caprotti: AR: title of painting and name of painters - we could introduce anonymous or unknown, then is no problem also with sparse records [1/16/13 2:49:45 PM] Olga Caprotti: AR: if there is a Picaso then it will be a new painter [1/16/13 2:51:08 PM] Olga Caprotti: MM: the question and answer grammar will be the same? [1/16/13 2:51:11 PM] Olga Caprotti: AR: yes [1/16/13 2:51:28 PM] Olga Caprotti: MM: if we could do the mapping from names to IDs [1/16/13 2:51:43 PM] Olga Caprotti: MM: queries will be easier [1/16/13 2:55:48 PM] Olga Caprotti: MD: we can have normalization done beforehand - [1/16/13 2:56:24 PM] Olga Caprotti: OC: for painter names - we could cross check different languages if records are given in multiple languages (which i guess they are generally not) [1/16/13 2:56:59 PM] Olga Caprotti: DD: about the book chapter, mariana can you please upload the paper to the website? [1/16/13 2:57:11 PM] Olga Caprotti: AR: or use svn please to deal with versions [1/16/13 2:57:24 PM] Olga Caprotti: AR: pls MOLTO SVN under wp8 [1/16/13 2:57:45 PM] Olga Caprotti: MD: mon tue next week for the book chapter [1/16/13 2:58:25 PM] Olga Caprotti: DD: next evaluation and tracking, we need to send texts, then get feedback, then carry out correction. [1/16/13 2:58:55 PM] Olga Caprotti: AR: evaluation will go to molto wp10 deliverable, done early for us to improve based onf eedback [1/16/13 2:59:13 PM] Olga Caprotti: AR: russian and polish will also be improved by the evaluation [1/16/13 3:00:51 PM] Olga Caprotti: MM: about YAQL, i was thinking to map some functors to specific SPARQL constructs. [1/16/13 3:01:14 PM] Olga Caprotti: AR: whatever works, for haskell it was easy because haskell supports same abstractions [1/16/13 3:01:58 PM] Olga Caprotti: MM: how should we proceed wth doing this? [1/16/13 3:02:29 PM] Olga Caprotti: AR: we must do it from YAQL to SPARQL, we do not have to do it in jan, but afterwards [1/16/13 3:02:44 PM] Olga Caprotti: AR: kramisir promised to help [1/16/13 3:03:19 PM] Olga Caprotti: AR: we can book some meetings between you MM and krasimir [1/16/13 3:03:42 PM] Olga Caprotti: AR: I was wondering that YAQL is now under wp8/grammars [1/16/13 3:04:30 PM] Olga Caprotti: AR: its proper place is wp4 [1/16/13 3:04:57 PM] Olga Caprotti: AR: if you read the README file it says to create a symbolic link [1/16/13 3:07:28 PM] Olga Caprotti: AR: if anyone has done any work in this local copy of YAQL, then it should be ported to the wp4 folder [1/16/13 3:09:00 PM] Olga Caprotti: DD: would like to see the mapping text on the ontotext site [1/16/13 3:09:27 PM] Olga Caprotti: MM: hope by the end of the week to be able to create a complete demo
NEF, JS, MC, LC, JvA, AR, OC
Apeldoorn not so appealing but MONET meeting is also taking place, Ütrecht or Tillburg.
Let Jeroen decide location. Time might be a issue because UPC starts the semester in the last 2 weeks also Norbert is away on vacation on first 2 weeks.
ACEwiki is at github. GF version of ACE: code is at grammaticalframework.org darcs repo. Let's not move them to MOLTO svn.
MOLTO semantic wiki -- New branch of ACEwiki in github. Development will be public.
https://github.com/Attempto/ACE-in-GF/tree/master/grammars/acewiki_aceowl - Grammars
https://github.com/Attempto/ACE-in-GF/blob/master/words/acewiki_aceowl/ - Test lexicon
Dynamically created articles
trees are linearized = an article is created
two views
articles written in different grammars - distinct spaces? scope?
GF version of ACE doesn't match coverage and precision of the current version of ACE 6.6. (Initial GF ACE was of ACE 6.0.)
(Size of ACE around half of RGL)
Problems in multilinguality: