Presentation is loading. Please wait.

Presentation is loading. Please wait.

ACBiMA: Advanced Chinese Bi-Character Word Morphological Analyzer 1 Ting-Hao (Kenneth) Huang Yun-Nung (Vivian) Chen Lingpeng Kong

Published byBathsheba Casey Modified over 9 years ago

Similar presentations

Presentation on theme: "ACBiMA: Advanced Chinese Bi-Character Word Morphological Analyzer 1 Ting-Hao (Kenneth) Huang Yun-Nung (Vivian) Chen Lingpeng Kong"— Presentation transcript:

1 ACBiMA: Advanced Chinese Bi-Character Word Morphological Analyzer 1 Ting-Hao (Kenneth) Huang Yun-Nung (Vivian) Chen Lingpeng Kong HTTP://ACBIMA.ORG

2 Outline ● Introduction ● Related Work ● Morphological Type Scheme ● Morphological Type Classification o Drived Word: Rule-Based Approach o Compond Word: ML Approach ● Experiments o ACBiMA Corpus 1.0 o Experimental Results ● Conclusion & Future Work 2:: ACBIMA.ORG ::

3 Outline ● Introduction ● Related Work ● Morphological Type Scheme ● Morphological Type Classification o Drived Word: Rule-Based Approach o Compond Word: ML Approach ● Experiments o ACBiMA Corpus 1.0 o Experimental Results ● Conclusion & Future Work 3:: ACBIMA.ORG ::

4 Introduction ● NLP tasks usually focus on segmented words ● Morphology is how words are composed with morphemes ● Usages of Chinese morphological structures o Sentiment Analysis (Ku, 2009; Huang, 2009) o POS Tagging (Qiu, 2008) o Word Segmentation (Gao, 2005) o Parsing (Li, 2011; Li, 2012; Zhang, 2013) ● Challenge for Chinese morphology o Lack of complete theories o Lack of category schema o Lack of toolkits 4:: ACBIMA.ORG :: 抗 菌 anti bacteria verb object

5 Outline ● Introduction ● Related Work ● Morphological Type Scheme ● Morphological Type Classification o Drived Word: Rule-Based Approach o Compond Word: ML Approach ● Experiments o ACBiMA Corpus 1.0 o Experimental Results ● Conclusion & Future Work 5:: ACBIMA.ORG ::

6 Related Work 6:: ACBIMA.ORG :: ● Focus on longer unknown words o Tseng, 2002; Tseng, 2005; Lu, 2008; Qiu, 2008 ● Focus on the functionality of morphemic characters o Bruno, 2010 ● Focus on Chinese bi-character words o Huang, et al., 2010 (LREC) o 52% multi-character Chinese tokens are bi-character o analyze Chinese morphological types o developed a suite of classifiers for type prediction Issue: covers only a subset of Chinese content words and has limited scalability

7 Outline ● Introduction ● Related Work ● Morphological Type Scheme ● Morphological Type Classification o Drived Word: Rule-Based Approach o Compond Word: ML Approach ● Experiments o ACBiMA Corpus 1.0 o Experimental Results ● Conclusion & Future Work 7:: ACBIMA.ORG ::

8 Morphological Type Scheme 8:: ACBIMA.ORG :: Chinese Bi-Char Content Word Single-Morpheme Word Synthetic Word Derived Word Compound Word dup, pfx, sfx, neg, ec a-head, conj, n-head, nsubj, v-head, vobj, vprt, els els

9 Derived Word 9:: ACBIMA.ORG ::

10 Compond Word 10:: ACBIMA.ORG ::

11 Outline ● Introduction ● Related Work ● Morphological Type Scheme ● Morphological Type Classification o Drived Word: Rule-Based Approach o Compond Word: ML Approach ● Experiments o ACBiMA Corpus 1.0 o Experimental Results ● Conclusion & Future Work 11:: ACBIMA.ORG ::

12 Morphological Type Classification 12:: ACBIMA.ORG :: ● Assumption: Chinese morphological structures are independent from word-level contexts (Tseng, 2002; Li, 2011) ● Derived words o Rule-based approach ● Compound words o ML-based approach

13 Derived Word: Rule-Based 13:: ACBIMA.ORG :: ● Idea o a morphologically derived word can be recognized based on its formation ● Approach o pattern matching rules ● Evaluation o Data: Chinese Treebank 7.0 o Result: o 2.9% of bi-char content words are annotated as derived words o Precision = 0.97 Rule-based methods are able to effectively recognizing derived words.

14 Compond Word: ML-Based 14:: ACBIMA.ORG :: ● Idea o The characteristics of individual characters can help decide the type of compond words ● ML classification models o Naïve Bayes o Random Forest o SVM

15 Classification Feature 15:: ACBIMA.ORG :: o Dict: Revised Mandarin Chinese Dictionary (MoE, 1994) o CTB: Chinese Treebank 5.1 (Xue et al., 2005)

16 Outline ● Introduction ● Related Work ● Morphological Type Scheme ● Morphological Type Classification o Drived Word: Rule-Based Approach o Compond Word: ML Approach ● Experiments o ACBiMA Corpus 1.0 o Experimental Results ● Conclusion & Future Work 16:: ACBIMA.ORG ::

17 ACBiMA Corpus 1.0 17:: ACBIMA.ORG :: ● Initial Set o 3,052 words o Extracted from CTB5 o Annotated with difficulty level ● Whole Set o 11,366 words o Initial Set + 3k words from CTB 5.1 + 6.5k words from (Huang, 2010)

18 Baseline Models 18:: ACBIMA.ORG :: 1)Majority 2)Stanford Dependency Map 3)Tabular Models o Step 1: assign the POS tags to each known character based on different heuristics o Step 2: assign the most frequent morphological type obtained from training data to each POS combination, e.g., “(VV, NN) = vobj”

19 Experimental Result 19:: ACBIMA.ORG :: o Setting: 10-fold cross-validation o Metrics: Macro F-measure (MF), Accuracy (ACC) Approachnsubj v- head a- head n- head vprtvobjconjelsMFACC Majority 000.5070000.172.340 Stanford Dep. Map 000.525.351.438.213.010.332.388 Tabular Stanford 0.2960.524.389.434.162.064.349.395 CTB.021.337.009.645.397.529.421.095.479.508 Dict 0.292.060.670.253.572.484.035.495.526 Tablular approaches perform better among all baselines.

20 Experimental Result 20:: ACBIMA.ORG :: o Setting: 10-fold cross-validation o Metrics: Macro F-measure (MF), Accuracy (ACC) Approachnsubj v- head a- head n- head vprtvobjconjelsMFACC Majority 000.5070000.172.340 Stanford Dep. Map 000.525.351.438.213.010.332.388 Tabular Stanford 0.2960.524.389.434.162.064.349.395 CTB.021.337.009.645.397.529.421.095.479.508 Dict 0.292.060.670.253.572.484.035.495.526 Naïve Base.273.406.195.523.679.566.547.188.519.518 Random Forest.250.421.063.760.803.643.656.076.647.674 SVM.413.541.288.748.791.657.636.271.662.665 ML-based methods outperform all baselines, where SVM & RF perform best.

21 Experimental Result 21:: ACBIMA.ORG :: o Setting: 10-fold cross-validation o Metrics: Macro F-measure (MF), Accuracy (ACC) Approachnsubj v- head a- head n- head vprtvobjconjelsMFACC Majority 000.5070000.172.340 Stanford Dep. Map 000.525.351.438.213.010.332.388 Tabular Stanford 0.2960.524.389.434.162.064.349.395 CTB.021.337.009.645.397.529.421.095.479.508 Dict 0.292.060.670.253.572.484.035.495.526 Naïve Base.273.406.195.523.679.566.547.188.519.518 Random Forest.250.421.063.760.803.643.656.076.647.674 SVM.413.541.288.748.791.657.636.271.662.665 Avg Difficulty 1.741.551.641.361.38 1.471.95--

22 Outline ● Introduction ● Related Work ● Morphological Type Scheme ● Morphological Type Classification o Drived Word: Rule-Based Approach o Compond Word: ML Approach ● Experiments o ACBiMA Corpus 1.0 o Experimental Results ● Conclusion & Future Work 22:: ACBIMA.ORG ::

23 Conclusion & Future Work 23:: ACBIMA.ORG :: ● Contribution o Linguistic  Propose a morphological type scheme  Develop a corpus containing about 11K words o Technical  Develop an effective morphological classifier o Practical  Data and tool available  Additional features for any Chinese task ● Future o Improve other NLP tasks by using ACBiMA

24 Q & A Thanks for your attentions!! 24 ● HTTP://ACBIMA.ORG

Download ppt "ACBiMA: Advanced Chinese Bi-Character Word Morphological Analyzer 1 Ting-Hao (Kenneth) Huang Yun-Nung (Vivian) Chen Lingpeng Kong"

Similar presentations

A Comparison of Implicit and Explicit Links for Web Page Classification Dou Shen 1 Jian-Tao Sun 2 Qiang Yang 1 Zheng Chen 2 1 Department of Computer Science.

A Comparison of Implicit and Explicit Links for Web Page Classification Dou Shen 1 Jian-Tao Sun 2 Qiang Yang 1 Zheng Chen 2 1 Department of Computer Science.
Document Summarization using Conditional Random Fields Dou Shen, Jian-Tao Sun, Hua Li, Qiang Yang, Zheng Chen IJCAI 2007 Hao-Chin Chang Department of Computer.

Document Summarization using Conditional Random Fields Dou Shen, Jian-Tao Sun, Hua Li, Qiang Yang, Zheng Chen IJCAI 2007 Hao-Chin Chang Department of Computer.
CS460/IT632 Natural Language Processing/Language Technology for the Web Lecture 2 (06/01/06) Prof. Pushpak Bhattacharyya IIT Bombay Part of Speech (PoS)

CS460/IT632 Natural Language Processing/Language Technology for the Web Lecture 2 (06/01/06) Prof. Pushpak Bhattacharyya IIT Bombay Part of Speech (PoS)
ICONIP 2005 Improve Naïve Bayesian Classifier by Discriminative Training Kaizhu Huang, Zhangbing Zhou, Irwin King, Michael R. Lyu Oct. 2005.

ICONIP 2005 Improve Naïve Bayesian Classifier by Discriminative Training Kaizhu Huang, Zhangbing Zhou, Irwin King, Michael R. Lyu Oct
Large-Scale Entity-Based Online Social Network Profile Linkage.

Large-Scale Entity-Based Online Social Network Profile Linkage.
计算机科学与技术学院 Chinese Semantic Role Labeling with Dependency-driven Constituent Parse Tree Structure Hongling Wang, Bukang Wang Guodong Zhou NLP Lab, School.

计算机科学与技术学院 Chinese Semantic Role Labeling with Dependency-driven Constituent Parse Tree Structure Hongling Wang, Bukang Wang Guodong Zhou NLP Lab, School.
TEMPLATE DESIGN © 2008 www.PosterPresentations.com Identifying Noun Product Features that Imply Opinions Lei Zhang Bing Liu Department of Computer Science,

TEMPLATE DESIGN © Identifying Noun Product Features that Imply Opinions Lei Zhang Bing Liu Department of Computer Science,
Recognizing Implicit Discourse Relations in the Penn Discourse Treebank Ziheng Lin, Min-Yen Kan, and Hwee Tou Ng Department of Computer Science National.

Recognizing Implicit Discourse Relations in the Penn Discourse Treebank Ziheng Lin, Min-Yen Kan, and Hwee Tou Ng Department of Computer Science National.
GENERATING AUTOMATIC SEMANTIC ANNOTATIONS FOR RESEARCH DATASETS AYUSH SINGHAL AND JAIDEEP SRIVASTAVA CS DEPT., UNIVERSITY OF MINNESOTA, MN, USA.

GENERATING AUTOMATIC SEMANTIC ANNOTATIONS FOR RESEARCH DATASETS AYUSH SINGHAL AND JAIDEEP SRIVASTAVA CS DEPT., UNIVERSITY OF MINNESOTA, MN, USA.
Predicting Text Quality for Scientific Articles Annie Louis University of Pennsylvania Advisor: Ani Nenkova.

Predicting Text Quality for Scientific Articles Annie Louis University of Pennsylvania Advisor: Ani Nenkova.
Shallow Processing: Summary Shallow Processing Techniques for NLP Ling570 December 7, 2011.

Shallow Processing: Summary Shallow Processing Techniques for NLP Ling570 December 7, 2011.
SUPPORT VECTOR MACHINES PRESENTED BY MUTHAPPA. Introduction Support Vector Machines(SVMs) are supervised learning models with associated learning algorithms.

SUPPORT VECTOR MACHINES PRESENTED BY MUTHAPPA. Introduction Support Vector Machines(SVMs) are supervised learning models with associated learning algorithms.
Predicting the Semantic Orientation of Adjective Vasileios Hatzivassiloglou and Kathleen R. McKeown Presented By Yash Satsangi.

Predicting the Semantic Orientation of Adjective Vasileios Hatzivassiloglou and Kathleen R. McKeown Presented By Yash Satsangi.
1 CSC 594 Topics in AI – Applied Natural Language Processing Fall 2009/2010 8. Shallow Parsing.

1 CSC 594 Topics in AI – Applied Natural Language Processing Fall 2009/ Shallow Parsing.
U.S. SENATE BILL CLASSIFICATION & VOTE PREDICTION Alessandra Paulino Rick Pocklington Serhat Selcuk Bucak.

U.S. SENATE BILL CLASSIFICATION & VOTE PREDICTION Alessandra Paulino Rick Pocklington Serhat Selcuk Bucak.
1/1/ Question Classification in English-Chinese Cross-Language Question Answering: An Integrated Genetic Algorithm and Machine Learning Approach Min-Yuh.

1/1/ Question Classification in English-Chinese Cross-Language Question Answering: An Integrated Genetic Algorithm and Machine Learning Approach Min-Yuh.
Presentation in IJCNN 2004 Biased Support Vector Machine for Relevance Feedback in Image Retrieval Hoi, Chu-Hong Steven Department of Computer Science.

Presentation in IJCNN 2004 Biased Support Vector Machine for Relevance Feedback in Image Retrieval Hoi, Chu-Hong Steven Department of Computer Science.
Sentence Classifier for Helpdesk Emails Anthony 6 June 2006 Supervisors: Dr. Yuval Marom Dr. David Albrecht.

Sentence Classifier for Helpdesk s Anthony 6 June 2006 Supervisors: Dr. Yuval Marom Dr. David Albrecht.
Parsing the NEGRA corpus Greg Donaker June 14, 2006.

Parsing the NEGRA corpus Greg Donaker June 14, 2006.
A Framework for Named Entity Recognition in the Open Domain Richard Evans Research Group in Computational Linguistics University of Wolverhampton UK

A Framework for Named Entity Recognition in the Open Domain Richard Evans Research Group in Computational Linguistics University of Wolverhampton UK

Similar presentations

Ads by Google