Tiwi and English/Structural transfer

From LING073
Jump to: navigation, search

Pre-Evaluation

Statistics about input files

Number of words in reference: 41

Number of words in test: 28

Number of unknown words (marked with a star) in test: 13

Percentage of unknown words: 14.63%

Results when removing unknown-word marks (stars)

Edit distance: 36

Word error rate (WER): 87.8%

Number of position-independent correct words: 5

Position-independent word error rate (PER): 87.8%

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

Edit distance: 36

Word error rate (WER): 87.8%

Number of position-independent correct words: 5

Position-independent word error rate (PER): 87.8%

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%

Examples

Gendered Nouns and Adjectives

Tiw-Eng

(tiw) ngarra kijinga → (eng) he is small

Tagger

(tiw) ngarra<prn><pers><p3><m><sg> kiji <adj><f> → (eng) prpers<prn><subj><p3><m><sg> be <vbser><pres><p3><sg> small <adj><sint>

Biltrans

^ngarra<prn><m><p3><sg>/he<prn><m><p3><sg>$ ^kiji<adj><f>/small<adj><sint><f>$^.<sent>/.<sent>$

Transfer

^he<prn><p3><m><sg>$ ^be<vbser><pres><p3><sg>$ ^small<adj><sint>$^.<sent>$

Insert "Be" Verbs

Tiw-Eng

(tiw) kamini naki? → (eng) what is this?

Tagger

(tiw) kami<prn><itg><m> naki<prn><dem><sg> → (eng) what <prn><itg><mf><sp> be <vbser><pres><p3><sg> this<prn><dem><mf><sg>

Biltrans

^kami<prn><itg><m>/what<prn><itg><m>/which<prn><itg><m>$ ^naki<prn><dem><sg>/this><prn><dem><sg>$^.<sent>/.<sent>$

Transfer

^what<prn><itg><m>$ ^be<vbser><pres><p3><sg>$ ^this><prn><dem><sg>$^.<sent>$

Word to Word Phrase

Tiw-Eng

(tiw) jupijupi awi kiyija mirrawu → (eng) soup and a little bit of tobacco

Tagger

(tiw) jupijupi<n> awi <cnjcoo> kiyija<prn><qnt> mirrawu <n> → (eng) soup<n><sg> and <cnjcoo> a little bit <adv> of <pr> tobacco <n><sg>

Biltrans

^jupijupi<n>/soup<n>$ ^awi<cnjcoo>/and<cnjcoo>$ ^kiyija<prn><qnt>/a little bit<adv>$ ^mirrawu<n>/tobacco<n>$^.<sent>/.<sent>$

Transfer

^soup<n><sg>$ ^and<cnjcoo>$ ^a little bit<adv>$ ^of<pr>$ ^tobacco<n><sg>$^.<sent>$

Post Evaluation

Statistics about input files

Number of words in reference: 41

Number of words in test: 28

Number of unknown words (marked with a star) in test: 13

Percentage of unknown words: 46.43%

Results when removing unknown-word marks (stars)

Edit distance: 36

Word error rate (WER): 87.8%

Number of position-independent correct words: 5

Position-independent word error rate (PER): 87.8%

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

Edit distance: 36

Word error rate (WER): 87.8%

Number of position-independent correct words: 5

Position-independent word error rate (PER): 87.8%

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%