Difference between revisions of "User:Pmodi/GSOC 2020 proposal: Hindi-Punjabi/progress"

From Apertium
Jump to navigation Jump to search
Line 55: Line 55:
! -
! -
! WER, PER On a set of 25 sentences(612 words)
! WER, PER On a set of 25 sentences(612 words)
|
|-





Revision as of 21:31, 8 June 2020

Current tasks(Week 1)

  • Fix some discrepancies due to postpositions(posttransfer)
  • Add Adverbs.
  • Clean up parallel text and compute WER.

Progress

Progress table

Week Stems Coverage WER,PER Progress
dates hin pan hin-pan hin pan hin-pan hin→pan pan→hin Evaluation Notes
1 May 1-24 - - - - 47% - - - - -
2 May 24-31 - +1400 - - 52% - 36.83%, 33.84% 40.03%, 36.93% - WER, PER On a set of 25 sentences(612 words)
3 June 1-7 - +500 - - 58% - 49.52%, 42.71% 48.65%, 44.03% - WER, PER On a set of 50 sentences(1038 words)

DONE

  • DONE - Added about 1400 adjective stems
  • DONE - Function words(cnj, det, prn, post, gen_endings), Coverage > 47%
  • DONE - Collected parallel texts to calculate WER,PER etc..
  • DONE - Added bidirectional dictionary(33k paradigms)
  • DONE - Fixed bidirectional translation i.e. pan->hin(gave close to human translation for small test set, even though similar transfer rules were copied)
  • DONE - Scraped all Wikipedia texts and made a combined frequency list.
  • DONE - Frequency lists using WikiExtractor on latest dump.

IN PROGRESS

  • IN PROGRESS - Adding Adverbs.
  • IN PROGRESS - Fix inconsistencies with parallel texts.
  • IN PROGRESS - Practicing basic understanding of Persian script.
  • IN PROGRESS - Extracting stems and suffix/prefix patterns using the frequency lists(might not be so beneficial since Punajbi and Hindi share inflections for different word forms)

TODO(Next week - Week 2)

  • Fix transfer rules for postpositions(fixing some issues).
  • Add verbs.
  • Get bidixes up-to-date with stems added to monodixes.

Literature(Apertium Wiki) Covered

  • DONE - Calculating Coverage.
  • DONE - A long introduction on Transfer Rules.
  • DONE - Transfer Rules examples
  • DONE - Wikipedia Dumps.
  • DONE - Generating Frequency lists.
  • DONE - Building Dictionaries#Monolingual.
  • DONE - Evaluation.
  • DONE - Extract.
  • DONE - Monodix Basics
  • DONE - Improved Corpus Based Paradigm Matching.
  • DONE - Transliteration.
  • DONE - Workflow reference.
  • DONE - Tagger Tranining.
  • DONE - Modes introduction.
  • DONE - Apertium-viewer.