Samest Project Plan

Estonian FST - 2013 - 2015

This comes before both Oahpa and MT implementation

• Revise the plamk fst or integrate it in the gt infra
  • Degree of adjustment of fst (discuss: Jaak, Heli, Trond + Sjur, soon after nov 18th)
  • Revision -- Tag adjustment
• Goals: Ability to generate Oahpa, MT, Dictionaries

Võru FST - 2013 - 2015

• Oahpa quality: generating the pedagogical lexicon
  • 70 stem classes, 2m to arrive at generating them

Work procedure

FST group now in the beginning, discussing both vro and est (Sulev, Jack, Jaak, Neeme, Heli, Trond, Heiki).

Goal:

• Select, make and proofread yaml files for testing (est and vro)
• Teach fst writing
• Learning by doing
• Later on leaving FST writing to the primary linguists

Extracting bilingual dictionaries

1. From corpora
   1. sme (gt) + fin (gt, hy) + est (filosoft) lemmatisation software in place
   2. corpora in place? Opus is in place (more sme-fin? Bible, other stuff from gt corpus?)
   3. TODO1: Set up Giza++ and Moses and do the alignment (Zürich)
   4. TODO2: Revise output with UPLUG + redo the tuning + find the best machine output (1+1 m)
   5. TODO3: Do the manual revision of the bilingual dictionaries (1+1 m)
2. From existing dictionaries
   1. (fin-rus/eng + rus/eng-est etc) (tasks 2.-4. 1 m)
   2. From WordNet, and from Wiktionary

Saami-Finnish MT

Alignment

Finnish-Estonian MT

The application said

• Evaluation of Oahpa by teachers and students - 2015-2016
• Publication of results at conferences - 2014 - 2016

Task order:

1. start with alignment
2. start with fst adjustment
3. start with getting some dictionary, e.g. nob-est, est-nob (additional nicety)

• Setting up Estonian and Võro Oahpa - 2014-1
• Numra est, vro 2014-1
• Leksa - 2014-1,2
• Morfa-S - 2014-1,2

Morfa-C - 2015

We will also give Vasta and Sahka a go

vro oahpa

1. set up 2014-1,
   Prototypes of Morfa-S, Leksa, Numra: 2014-2
   Morfa-s: fst to generate forms of vocabulary
   Leksa: vocabulary + semantic markup
2. Use in courses, feedback, adjustment: 2014-3,4

Later: Morfa-c, further work

Teachers

• est: Ilona and others
• vro: Sulev is the teacher + collegues in Võro Institute + TÜ

Time schedule:

• 2013-4 Estonian fst (discussion, adjustment) , Võro fst: approaching oahpa quality
• 2014-1 Set up Oahpa for est, vro
• 2014-2 Work with Oahpa for est, vro; work with fsts
• 2014-3 Oahpa in courses: vro, est
• 2014-4 Oahpa in courses: vro, est
• 2015 It will be planned later
• 2016 It will be planned later

Time - month schedule

• Weighting, tasks: EstFST + VroFST + Oahpa + MT + Admin = as above
• Weighting, years: 2014, 2015, 2016 = quite even

People

Meetings

Startup in Tromsø in january/february 2014 Meetings in Tartu Mid-project in

Other expenses Laptops 2000 3 6000 Travel 2000 13 31200 Conferences 5000 4 24000 61200

Workshops

ICALL workshop on Oahpa-related things in cooperation with Tromsø ICALL group

Brainstorming:

• The "See what I have done" - paper
  • Võru fst
  • Estonian Oahpa
  • MT comparision RBMT vs. SMT
• Specific problems papers
  • cross-language FST comparisons
  • Papers based upon user logs
  • Papers based upon specific problems during our work