Page 168 - My FlipBook
P. 168
研
人 Associate Research Fellow
員 馬偉雲 Wei-Yun Ma
Faculty Ph.D., Computer Science, Columbia University, United States
T +886-2-2788-3799 ext. 1819 E ma@iis.sinica.edu.tw
F +886-2-2782-4814 W www.iis.sinica.edu.tw/pages/ma
・ Assistant Research Fellow, Institute of Information Science, Academia Sinica (2014-2020)
・ Ph.D., Computer Science, Columbia University (2014)
・ Top3 in the rst Formosa Grand Challenge-Talk to AI (2019)
・ PIXNET best product award for Chatbot (2017)
・ Best paper award in IALP (2017)
・ Best Demo - Special Mention Award at WWW (2017)
Research Description
My research interests include Natural Language Processing (NLP), Machine Learning, Deep Learning, Knowledge Graph, Dialog Agent and
Question Answering. I am fascinated by the idea of giving computers the ability to understand human language.
Recently I have mainly focused on NLP problems concerning knowledge-based information processing, a process which is strongly motivated
by the deluge of information available on the Internet and knowledge bases. To develop a practical and high-quality knowledge base for the
need of knowledge-based information processing is very critical and also challenging. The number of entities can be millions, ten millions,
and even endless. It is impossible to build manually, so it must be built in an automated way. At the same time, its inference mechanism,
including causality, relationships, action processes, etc. must also be considered together, so the representation form and the representation
ability of such knowledge base also needs to be carefully designed.
To build such a practical knowledge base, I am currently engaged in research on knowledge acquisition, representation, reasoning and
utilization. I am also fascinated by addressing NLP tasks by utilizing deep reinforcement learning models, where the models concern with
how software agents ought to take actions in an environment so as to maximize rewards.
Publications on Asian Language Processing (IALP), December 2017, (received
Best Paper, Top1 out of 96 accepted papers)
1. Mu Yang, Chi-Yen Chen, Yi-Hui Lee, Qian-Hui Zeng, and Wei-
Yun Ma, "Headword-Oriented Entity Linking: A New Entity 10. Peng-Hsuan Li, Ruo-Ping Dong, Yu-SiangWang, Ju-Chieh Chou,
Linking Task with Dataset and Baseline," LREC, May 2020 Wei-Yun Ma, "Leveraging Linguistic Structures for Named Entity
Recognition with Bidirectional Recursive Neural Networks,"
2. Peng-Hsuan Li, Tsan-Yu Yang, and Wei-Yun Ma, "CA-EHN: International Conference on EMNLP, September 2017.
Commonsense Analogy from E-HowNet," LREC, May 2020
11. Peng-Yu Chen, Yi-Hui Lee, Yueh-Han Wu, Wei-Yun Ma, "IExM:
3. Peng-Hsuan Li, Tsu-Jui Fu, and Wei-Yun Ma, "Why Attention? Information Extraction System for Movies," WWW 2017 Demo
Analyze BiLSTM Deficiency and Its Remedies in the Case of Track (receive Best Demo - Special Mention Award), April 2017.
NER," International Conference on AAAI, Feb 2020.
12. Hsin-Yang Wang, Wei-Yun Ma, "Integrating Semantic Knowledge
4. Lifeng Jin, Linfeng Song, Yue Zhang, Kun Xu, Wei-Yun Ma, into Lexical Embeddings Based on Information Content
Dong Yu, "Relation Extraction Exploiting Full Dependency Measurement," International Conference on EACL, April 2017.
Forests," International Conference on AAAI, Feb 2020.
13. Hsin-Yang Wang, Wei-Yun Ma, "CKIP Valence-Arousal Predictor
5. Tsu-Hui Fu, Wei-Yun Ma, "GraphRel: Modeling Text as for IALP 2016 Shared Task," International Conference on Asian
Relational Graphs for Joint Entity and Relation Extraction," Language Processing (IALP) (Top 1 out of 22 international teams
International Conference on ACL, July 2019. on Valence of IALP 2016 Sentiment Task), November 2016.
6. Tsu-Hui Fu, Wei-Yun Ma, "Speed Reading: Learning to Read 14. Wei-Yun Ma and Kathleen McKeown, "System Combination
ForBackward via Shuttle," International Conference on EMNLP, for Machine Translation through Paraphrasing," Proceedings
October 2018. of Conference on Empirical Methods in Natural Language
Processing (EMNLP), September 2015.
7. Wei-Yun Ma and Yueh-Yin Shih, "Extended HowNet 2.0 - An
Entity-Relation Common-Sense Representation Model," LREC, 15. Wei-Yun Ma and Kathleen McKeown, "Where's the Verb
May 2018. Correcting Machine Translation During Question Answering,"
Proceedings of ACL-IJCNLP, July 2009.
8. Chi-Yen Chen, Wei-Yun Ma, "Word Embedding Evaluation
Datasets and Wikipedia Title Embedding for Chinese," LREC ,
May 2018.
9. Chi-Yen Chen, Wei-Yun Ma, "Embedding Wikipedia Title Based
on Its Wikipedia Text and Categories," International Conference
166
人 Associate Research Fellow
員 馬偉雲 Wei-Yun Ma
Faculty Ph.D., Computer Science, Columbia University, United States
T +886-2-2788-3799 ext. 1819 E ma@iis.sinica.edu.tw
F +886-2-2782-4814 W www.iis.sinica.edu.tw/pages/ma
・ Assistant Research Fellow, Institute of Information Science, Academia Sinica (2014-2020)
・ Ph.D., Computer Science, Columbia University (2014)
・ Top3 in the rst Formosa Grand Challenge-Talk to AI (2019)
・ PIXNET best product award for Chatbot (2017)
・ Best paper award in IALP (2017)
・ Best Demo - Special Mention Award at WWW (2017)
Research Description
My research interests include Natural Language Processing (NLP), Machine Learning, Deep Learning, Knowledge Graph, Dialog Agent and
Question Answering. I am fascinated by the idea of giving computers the ability to understand human language.
Recently I have mainly focused on NLP problems concerning knowledge-based information processing, a process which is strongly motivated
by the deluge of information available on the Internet and knowledge bases. To develop a practical and high-quality knowledge base for the
need of knowledge-based information processing is very critical and also challenging. The number of entities can be millions, ten millions,
and even endless. It is impossible to build manually, so it must be built in an automated way. At the same time, its inference mechanism,
including causality, relationships, action processes, etc. must also be considered together, so the representation form and the representation
ability of such knowledge base also needs to be carefully designed.
To build such a practical knowledge base, I am currently engaged in research on knowledge acquisition, representation, reasoning and
utilization. I am also fascinated by addressing NLP tasks by utilizing deep reinforcement learning models, where the models concern with
how software agents ought to take actions in an environment so as to maximize rewards.
Publications on Asian Language Processing (IALP), December 2017, (received
Best Paper, Top1 out of 96 accepted papers)
1. Mu Yang, Chi-Yen Chen, Yi-Hui Lee, Qian-Hui Zeng, and Wei-
Yun Ma, "Headword-Oriented Entity Linking: A New Entity 10. Peng-Hsuan Li, Ruo-Ping Dong, Yu-SiangWang, Ju-Chieh Chou,
Linking Task with Dataset and Baseline," LREC, May 2020 Wei-Yun Ma, "Leveraging Linguistic Structures for Named Entity
Recognition with Bidirectional Recursive Neural Networks,"
2. Peng-Hsuan Li, Tsan-Yu Yang, and Wei-Yun Ma, "CA-EHN: International Conference on EMNLP, September 2017.
Commonsense Analogy from E-HowNet," LREC, May 2020
11. Peng-Yu Chen, Yi-Hui Lee, Yueh-Han Wu, Wei-Yun Ma, "IExM:
3. Peng-Hsuan Li, Tsu-Jui Fu, and Wei-Yun Ma, "Why Attention? Information Extraction System for Movies," WWW 2017 Demo
Analyze BiLSTM Deficiency and Its Remedies in the Case of Track (receive Best Demo - Special Mention Award), April 2017.
NER," International Conference on AAAI, Feb 2020.
12. Hsin-Yang Wang, Wei-Yun Ma, "Integrating Semantic Knowledge
4. Lifeng Jin, Linfeng Song, Yue Zhang, Kun Xu, Wei-Yun Ma, into Lexical Embeddings Based on Information Content
Dong Yu, "Relation Extraction Exploiting Full Dependency Measurement," International Conference on EACL, April 2017.
Forests," International Conference on AAAI, Feb 2020.
13. Hsin-Yang Wang, Wei-Yun Ma, "CKIP Valence-Arousal Predictor
5. Tsu-Hui Fu, Wei-Yun Ma, "GraphRel: Modeling Text as for IALP 2016 Shared Task," International Conference on Asian
Relational Graphs for Joint Entity and Relation Extraction," Language Processing (IALP) (Top 1 out of 22 international teams
International Conference on ACL, July 2019. on Valence of IALP 2016 Sentiment Task), November 2016.
6. Tsu-Hui Fu, Wei-Yun Ma, "Speed Reading: Learning to Read 14. Wei-Yun Ma and Kathleen McKeown, "System Combination
ForBackward via Shuttle," International Conference on EMNLP, for Machine Translation through Paraphrasing," Proceedings
October 2018. of Conference on Empirical Methods in Natural Language
Processing (EMNLP), September 2015.
7. Wei-Yun Ma and Yueh-Yin Shih, "Extended HowNet 2.0 - An
Entity-Relation Common-Sense Representation Model," LREC, 15. Wei-Yun Ma and Kathleen McKeown, "Where's the Verb
May 2018. Correcting Machine Translation During Question Answering,"
Proceedings of ACL-IJCNLP, July 2009.
8. Chi-Yen Chen, Wei-Yun Ma, "Word Embedding Evaluation
Datasets and Wikipedia Title Embedding for Chinese," LREC ,
May 2018.
9. Chi-Yen Chen, Wei-Yun Ma, "Embedding Wikipedia Title Based
on Its Wikipedia Text and Categories," International Conference
166
