User:Oğuz/GSoC 2019 progress

From Apertium
Revision as of 11:34, 28 June 2019 by Oğuz (talk | contribs)
Jump to navigation Jump to search

Progress on 2019 GSoC Project "http://wiki.apertium.org/wiki/User:Oğuz/GSoC_2019[Turkic MT improvement"].




WER results

1st evaluation WER results:


Uzbek


Test file: 'istanbultr.txt' Reference file 'turistanbul.txt'

Statistics about input files


Number of words in reference: 206 Number of words in test: 208 Number of unknown words (marked with a star) in test: 28 Percentage of unknown words: 13.46 %

Results when removing unknown-word marks (stars)


Edit distance: 78 Word error rate (WER): 37.86 % Number of position-independent correct words: 132 Position-independent word error rate (PER): 36.89 %

Results when unknown-word marks (stars) are not removed


Edit distance: 76 Word Error Rate (WER): 36.89 % Number of position-independent correct words: 134 Position-independent word error rate (PER): 35.92 %

Statistics about the translation of unknown words


Number of unknown words which were free rides: -2 Percentage of unknown words that were free rides: -7.14 %


Kyrgyz


Test file: 'kazantr.txt' Reference file 'kazanturkce.txt'

Statistics about input files


Number of words in reference: 223 Number of words in test: 227 Number of unknown words (marked with a star) in test: 55 Percentage of unknown words: 24.23 %

Results when removing unknown-word marks (stars)


Edit distance: 113 Word error rate (WER): 50.67 % Number of position-independent correct words: 119 Position-independent word error rate (PER): 48.43 %

Results when unknown-word marks (stars) are not removed


Edit distance: 108 Word Error Rate (WER): 48.43 % Number of position-independent correct words: 124 Position-independent word error rate (PER): 46.19 %

Statistics about the translation of unknown words


Number of unknown words which were free rides: -5 Percentage of unknown words that were free rides: -9.09 %


Tatar


Test file: 'kazantr.txt' Reference file 'kazantur.txt'

Statistics about input files


Number of words in reference: 195 Number of words in test: 210 Number of unknown words (marked with a star) in test: 36 Percentage of unknown words: 17.14 %

Results when removing unknown-word marks (stars)


Edit distance: 103 Word error rate (WER): 52.82 % Number of position-independent correct words: 112 Position-independent word error rate (PER): 50.26 %

Results when unknown-word marks (stars) are not removed


Edit distance: 102 Word Error Rate (WER): 52.31 % Number of position-independent correct words: 113 Position-independent word error rate (PER): 49.74 %

Statistics about the translation of unknown words


Number of unknown words which were free rides: -1 Percentage of unknown words that were free rides: -2.78 %


Uyghur


Test file: 'cumhuriyet-1.txt' Reference file 'cumhuriyetturkce.txt'

Statistics about input files


Number of words in reference: 354 Number of words in test: 359 Number of unknown words (marked with a star) in test: 20 Percentage of unknown words: 5.57 %

Results when removing unknown-word marks (stars)


Edit distance: 61 Word error rate (WER): 17.23 % Number of position-independent correct words: 299 Position-independent word error rate (PER): 16.95 %

Results when unknown-word marks (stars) are not removed


Edit distance: 61 Word Error Rate (WER): 17.23 % Number of position-independent correct words: 299 Position-independent word error rate (PER): 16.95 %

Statistics about the translation of unknown words


Number of unknown words which were free rides: 0 Percentage of unknown words that were free rides: 0.00 %