Baseline corpus for the evaluation of phrasebook

0
ID: 
10.8
Task leader: 
olga.caprotti
Assignees: 
aarne.ranta
Assignees: 
olga.caprotti
Assignees: 
ramona.enache
Status: 
Accepted
Timeframe: 
Nov 2012 - Dec 2012

Jussi, from UHEL WP, needs some material to design the evaluation process. As we are piloting with the tourist phrasebook, its material should be sent as soon as possible. It must be something to base the evaluation on, so even just seeing the translatable text would help a lot.

In order of preference:

  • All language pairs of the tourist corpus (EN-XX)
  • Some language pairs
  • A single language pair (EN-DE for example)
  • Just the English phrases

A few hundred units to look at - I have no idea about the volume at this moment. The format of the material is irrelevant at this point - just a simple text file will do.