Presentation is loading. Please wait.

Presentation is loading. Please wait.

Exploiting Shared Chinese Characters in Chinese Word Segmentation Optimization for Chinese-Japanese Machine Translation Chenhui Chu, Toshiaki Nakazawa,

Similar presentations


Presentation on theme: "Exploiting Shared Chinese Characters in Chinese Word Segmentation Optimization for Chinese-Japanese Machine Translation Chenhui Chu, Toshiaki Nakazawa,"— Presentation transcript:

1 Exploiting Shared Chinese Characters in Chinese Word Segmentation Optimization for Chinese-Japanese Machine Translation Chenhui Chu, Toshiaki Nakazawa, Daisuke Kawahara, Sadao Kurohashi Graduate School of Informatics, Kyoto University EAMT2012 (2012/05/28) 1

2 Outline Word Segmentation Problems Common Chinese Characters Chinese Word Segmentation Optimization Experiments Discussion Related Work Conclusion and Future Work 2

3 Outline Word Segmentation Problems Common Chinese Characters Chinese Word Segmentation Optimization Experiments Discussion Related Work Conclusion and Future Work 3

4 Word Segmentation for Chinese-Japanese MT 4 Mr. Kosaka is the founder of The Japan Society for Clinical Anesthesiologists. 小坂 / 先生 / は / 日本 / 臨床 / 麻酔 / 学会 / の / 創始 / 者 / である / 。 小 / 坂 / 先生 / 是 / 日本 / 临床 / 麻醉 / 学会 / 的 / 创始者 / 。 Zh: Ja: Ref: 小坂先生は日本臨床麻酔学会の創始者である。 小坂先生是日本临床麻醉学会的创始者。 Zh: Ja:

5 Word Segmentation Problems in Chinese-Japanese MT Unknown words – Affect segmentation accuracy and consistency Word segmentation granularity – Affect word alignment 5 小坂 / 先生 / は / 日本 / 臨床 / 麻酔 / 学会 / の / 創始 / 者 / である / 。 小 / 坂 / 先生 / 是 / 日本 / 临床 / 麻醉 / 学会 / 的 / 创始者 / 。 Mr. Kosaka is the founder of The Japan Society for Clinical Anesthesiologists. Zh: Ja: Ref:

6 Outline Introduction Common Chinese Characters Chinese Word Segmentation Optimization Experiments Discussion Related Work Conclusion and Future Work 6

7 Chinese Characters Chinese characters are used both in Chinese (Hanzi) and Japanese (Kanji) There are many common Chinese characters between Hanzi and Kanji We made a common Chinese characters mapping table for 6,355 JIS Kanji (Chu et al., 2012) 7

8 Common Chinese Characters Related Studies Automatic sentence alignment task (Tan et al., 1995) Dictionary construction (Goh et al., 2005) Word level semantic relations investigation (Huang et al., 2008) Phrase alignment (Chu et al., 2011) This study exploits common Chinese characters in Chinese word segmentation optimization for MT 8

9 Outline Word Segmentation Problems Common Chinese Characters Chinese Word Segmentation Optimization Experiments Discussion Related Work Conclusion and Future Work 9

10 Reason for Chinese Word Segmentation Optimization Segmentation for Japanese is easier than Chinese, because Japanese uses Kana other than Chinese characters F-score for Japanese segmentation is nearly 99% (Kudo et al., 2004), while that for Chinese is still about 95% (Wang et al., 2011) Therefore, we only do word segmentation optimization for Chinese, and keep the Japanese segmentation results 10

11 11 Parallel Training Corpus ① Chinese Lexicons Extraction Chinese Annotated Corpus for Chinese Segmenter ② Chinese Lexicons Incorporation ③ Short Unit Transformation Optimized Chinese Segmenter Training Short Unit Chinese Corpus for Chinese Segmenter Common Chinese Characters Chinese Lexicons System Dictionary of Chinese Segmenter System Dictionary with Chinese Lexicons System Dictionary with Chinese Lexicons

12 12 Parallel Training Corpus ① Chinese Lexicons Extraction Chinese Annotated Corpus for Chinese Segmenter ② Chinese Lexicons Incorporation ③ Short Unit Transformation Optimized Chinese Segmenter Training Short Unit Chinese Corpus for Chinese Segmenter Common Chinese Characters Chinese Lexicons System Dictionary of Chinese Segmenter System Dictionary with Chinese Lexicons System Dictionary with Chinese Lexicons

13 ① Chinese Lexicons Extraction Step 1: Segment Chinese and Japanese sentences in the parallel training corpus Step 2: Convert Japanese Kanji tokens into Chinese using the mapping table we made (Chu et al., 2012) Step 3: Extract the converted tokens as Chinese lexicons if they exist in the corresponding Chinese sentence 13

14 Extraction Example 14 小坂 / 先生 / は / 日本 / 臨床 / 麻酔 / 学会 / の / 創始 / 者 / である / 。 小 / 坂 / 先生 / 是 / 日本 / 临床 / 麻醉 / 学会 / 的 / 创始 者 / 。 Mr. Kosaka is the founder of The Japan Society for Clinical Anesthesiologists. Zh: Ja: Ref: 小坂 / 先生 / は / 日本 / 临床 / 麻醉 / 学会 / の / 创始 / 者 / である / 。 Ja: Kanji tokens conversion Check 小坂 先生 日本 临床 麻醉 学会 创始 者 Chinese Lexicons : Extraction

15 15 Parallel Training Corpus ① Chinese Lexicons Extraction Chinese Annotated Corpus for Chinese Segmenter ② Chinese Lexicons Incorporation ③ Short Unit Transformation Optimized Chinese Segmenter Training Short Unit Chinese Corpus for Chinese Segmenter Common Chinese Characters Chinese Lexicons System Dictionary of Chinese Segmenter System Dictionary with Chinese Lexicons System Dictionary with Chinese Lexicons

16 ② Chinese Lexicons Incorporation Using a system dictionary is helpful for Chinese word segmentation (Low et al., 2005; Wang et al., 2011) We incorporate the extracted lexicons into the system dictionary of a Chinese segmenter Assign POS tags by converting the POS tags assigned by the Japanese segmenter using POS tags mapping table between Chinese and Japanese 16

17 17 CTB (Penn Chinese Treebank)JUMAN (Kurohashi et al., 1994) AD(adverb) 副詞 (adverb) CC(coordinating conjunction) 接続詞 (conjunction) CD(cardinal number) 名詞 (noun)[ 数詞 (numeral noun)] FW(foreign words) 未定義語 (undefined word)[ アルファベット (alphabet)] IJ(interjection) 感動詞 (interjection) M(measure word) 接尾辞 (suffix)[ 名詞性名詞助数辞 (measure word suffix)] NN(common noun) 名詞 (noun)[ 普通名詞 (common noun)/ サ変名詞 (sahen noun)/ 形式名詞 (formal noun)/ 副詞的名詞 (adverbial noun)], 接尾辞 (suffix)[ 名詞性名詞接尾辞 (noun suffix)/ 名詞性特殊接尾辞 (special noun suffix)] NR(proper noun) 名詞 (noun)[ 固有名詞 (proper noun)/ 地名 (place name)/ 人名 (person name)/ 組織名 (organization name)] NT(temporal noun) 名詞 (noun)[ 時相名詞 (temporal noun)] PU(punctuation) 特殊 (special word) VA(predicative adjective) 形容詞 (adjective) VV(other verb) 動詞 (verb)/ 名詞 (noun)[ サ変名詞 (sahen noun)]

18 18 Parallel Training Corpus ① Chinese Lexicons Extraction Chinese Annotated Corpus for Chinese Segmenter ② Chinese Lexicons Incorporation ③ Short Unit Transformation Optimized Chinese Segmenter Training Short Unit Chinese Corpus for Chinese Segmenter Common Chinese Characters Chinese Lexicons System Dictionary of Chinese Segmenter System Dictionary with Chinese Lexicons System Dictionary with Chinese Lexicons

19 ③ Short Unit Transformation Adjusting Chinese word segmentation to make tokens 1-to-1 mapping as many as possible between a parallel sentences can improve alignment accuracy (Bai et al., 2008) Wang et al. (2010) proposed a short unit standard for Chinese word segmentation, which can reduce the number of 1-to-n alignments and improve MT performance 19

20 Our Method We transform the annotated training data of Chinese segmenter utilizing the extracted lexicons 20 从 _P/ 有效性 _NN / 高 _VA/ 的 _DEC/ 格要素 _NN /… 从 _P/ 有效 _NN/ 性 _NN / 高 _VA/ 的 _DEC/ 格 _NN/ 要素 _NN /… Short: CTB: Lexicon: 有效 (effective)Lexicon : 要素 (element) From case element with high effectiveness … Ref:

21 Constraints We do not use the extracted lexicons that are composed of only one Chinese character 21 歌 (song) ・・・ 歌颂 (praise) 歌 (song)/ 颂 (praise) long tokenextracted lexiconsshort unit tokens

22 Outline Word Segmentation Problems Common Chinese Characters Chinese Word Segmentation Optimization Experiments Discussion Related Work Conclusion and Future Work 22

23 Two Kinds of Experiments Experiments on Moses Experiments on Kyoto example-based machine translation (EBMT) system Nakazawa and Kurohashi, 2011) – A dependency tree-based decoder 23

24 Experimental Settings on MOSES (1/2) 24 Parallel Training CorpusZh-Ja paper abstract corpus (680k sentences) Chinese Annotated CorpusNICT Chinese Treebank (same domain of parallel corpus, 9,792 sentences) CTB 7 (31,131 sentences) Chinese SegmenterIn-house Chinese segmenter Japanese SegmenterJUMAN (Kurohashi et al., 1994) MT systemMoses with default options, except for the distortion limit (6→20)

25 Experimental Settings on MOSES (2/2) Baseline: Only using the lexicons extracted from Chinese annotated corpus Incorporation: Incorporate the extracted Chinese lexicons Short unit: Incorporate the extracted Chinese lexicons and train the Chinese segmenter on the short unit training data 25

26 Results of Chinese-to-Japanese Translation Experiments on MOSES BLEUNICT Chinese TreebankCTB 7 Baseline34.9236.64 Incorporation36.1937.39 Short unit36.8238.59 26 CTB 7 shows better performance because the size is more than 3 times of NICT Chinese Treebank Lexicons extracted from a paper abstract domain also work well on other domains (i.e. CTB 7)

27 Results of Japanese-to-Chinese Translation Experiments on MOSES 27 BLEUNICT Chinese TreebankCTB 7 Baseline31.4231.83 Incorporation31.2432.34 Short unit31.95 Not significant compared to Zh-to-Ja, because our proposed approach does not change the segmentation results of input Japanese sentences

28 Experimental Settings on EBMT (1/2) 28 Parallel Training CorpusZh-Ja paper abstract corpus (680k sentences) Chinese Annotated CorpusCTB 7 (31,131 sentences) Chinese SegmenterIn-house Chinese segmenter Japanese SegmenterJUMAN (Kurohashi et al., 1994) MT systemKyoto example-based machine translation (EBMT) system

29 Experimental Settings on EBMT (2/2) Baseline: Only using the lexicons extracted from Chinese annotated corpus Short unit: Incorporate the extracted Chinese lexicons and train the Chinese segmenter on the short unit training data 29

30 Results of Translation Experiments on EBMT 30 BLEUZh-to-JaJa-to-Zh Baseline22.8419.10 Short unit23.3619.15 Translation performance is worse than MOSES, because EBMT suffers from low accuracy of Chinese parser Improvement by short unit is not significant because the Chinese parser is not trained on short unit segmented training data

31 Outline Word Segmentation Problems Common Chinese Characters Chinese Word Segmentation Optimization Experiments Discussion Related Work Conclusion and Future Work 31

32 Short Unit Effectiveness on MOSES 32 Baseline (BLEU=49.38) Input: 本 / 论文 / 中 / , / 提议 / 考虑 / 现存 / 实现 / 方式 / 的 / 功能 / 适应性 / 决定 / 对策 / 目标 / 的 / 保密 / 基本 / 设计法 / 。 Output: 本 / 論文 / で / は / , / 提案 / する / 適応 / 的 / 対策 / を / 決定 / する / セキュリティ / 基 本 / 設計 / 法 / を / 考える / 現存 / の / 実現 / 方式 / の / 機能 / を / 目標 / と / して / いる / . Short unit (BLEU=56.33) Input: 本 / 论文 / 中 / , / 提议 / 考虑 / 现存 / 实现 / 方式 / 的 / 功能 / 适应 / 性 / 决定 / 对策 / 目标 / 的 / 保密 / 基本 / 设计 / 法 / 。 Output: 本 / 論文 / で / は / , / 提案 / する / 考え / 現存 / の / 実現 / 方式 / の / 機能 / 的 / 適応 / 性 / を / 決定 / する / 対策 / 目標 / の / セキュリティ / 基本 / 設計 / 法 / を / 提案 / する / . Reference 本 / 論文 / で / は / , / 対策 / 目標 / を / 現存 / の / 実現 / 方式 / の / 機能 / 的 / 適合 / 性 / も / 考慮 / して / 決定 / する / セキュリティ / 基本 / 設計 / 法 / を / 提案 / する / . (In this paper, we propose a basic security design method also consider functional suitability of the existing implementation method for determining countermeasures target.)

33 Number of Extracted Lexicons 33 SourceNICT TreebankCTB 7 Chinese annotated corpus13,47126,202 Short unit Chinese corpus12,62725,490 Source Parallel training corpus18,584 The number of extracted lexicons deceased after short unit transformation because duplicated lexicons increased

34 Short Unit Transformation Percentage NICT Chinese Treebank – 6,623 tokens out of 257,825 been transformed to 13,469 short unit tokens, the percentage is 2.57% CTB 7 – 19,983 tokens out of 718,716 been transformed to 41,336 short unit tokens, the percentage is 2.78% 34

35 Short Unit Transformation Problems (1/3) Improper transformation problem 35 好意 (favor) ・・・ 不好意思 (sorry) 不 (not)/ 好意 (favor)/ 思 (think) long tokenextracted lexiconsshort unit tokens

36 Short Unit Transformation Problems (2/3) 36 充电 (charge) 电器 (electric equipment) ・・・ 充电器 (charger) 充 (charge)/ 电器 (electric equipment) 充电 (charge)/ 器 (device) long tokenextracted lexiconsshort unit tokens Transformation ambiguity problem

37 Short Unit Transformation Problems (3/3) 37 实验 (test) ・・・ 被实验者 _NN (test subject) 被 _NN(be)/ 实验 _NN(test)/ 者 _NN(person) long tokenextracted lexiconsshort unit tokens The correct POS tag for “ 被 (be)” should be LB (“ 被 ” in long bei-const) POS Tag assignment problem

38 Outline Word Segmentation Problems Common Chinese Characters Chinese Word Segmentation Optimization Experiments Discussion Related Work Conclusion and Future Work 38

39 Bai et al., 2008 Proposed a method of learning affix rules from a aligned Chinese-English bilingual terminology bank to adjust Chinese word segmentation in the parallel corpus directly 39

40 Wang et al., 2010 Proposed a method based on transfer rules and a transfer database. – The transfer rules are extracted from alignment results of annotated Chinese and segmented Japanese training data – The transfer database is constructed using external lexicons, and is manually modified 40

41 Conclusion We proposed an approach of exploiting common Chinese characters in Chinese word segmentation optimization for Chinese- Japanese MT Experimental results of Chinese-Japanese MT on a phrase-based SMT system indicated that our approach can improve MT performance significantly 41

42 Future Work Solve the short unit transformation problems Evaluate the proposed approach on parallel corpus of other domains 42


Download ppt "Exploiting Shared Chinese Characters in Chinese Word Segmentation Optimization for Chinese-Japanese Machine Translation Chenhui Chu, Toshiaki Nakazawa,"

Similar presentations


Ads by Google