Page 64 - profile2014.indd

助研究員
古倫維 Lun-Wei Ku

Assistant Research Fellow
Ph.D., Computer Science, National Taiwan University
Tel: +886-2-2788-3799 ext.1808 Fax: +886-2-2782-4814
Email: lwku@iis.sinica.edu.tw
http://www.iis.sinica.edu.tw/pages/lwku

● Organizer, SocialNLP workshop, Research Description
AFNLP (2012-present)
● NSC excellent junior research My research interests include natural language processing (NLP), computational linguistics, in-
investigator, three year grant formation retrieval, and in particular, sentiment analysis and opinion mining. My approaches
(2012-2015) focus on how to utilize linguistic rules, common sense, and real world knowledge, such as syn-
tactic/semantic cues, thesaurus, ontology, and information acquired from explosive web data, to
● Good Design Award (G-Mark), understand natural languages. I have also developed the rst Chinese opinion analysis system,

Selected: IlluMe, Japan (2011) CopeOpi. Recently, I am interested in deep understanding, emotion detection, and social media
● Assistant Professor, NYUST, CSIE, text analysis. Related research topics include:
Taiwan (2011-2012) Semantic Role Labeling: We try to enhance the performance of semantic role labeling in the
Sinica Parser. We utilize real world knowledge from E-Hownet together with contextual features
● ROCLING Distinct Doctorial Dis- on several models and their combinations. Techniques are also applied to sentiment analysis,
sertation Award
entailment, and question answering.
● IBM Ph.D. Fellowship
Readers’ Interest Analysis: We are interested in predicting readers’ interest after their reading.

From di erent dimensions, including semantic, physical, and sentiment aspects, we pursue
terms representing interests instead of topics.
Emotion Detection and Representation: We are interested in the emotion visualization of texts. We have analyzed a large quantity of social
media posts and their emotion tags. For deep understanding, we have developed an automatic emotion pattern extraction approach to detect

emotions. Then, we used emotion colors de ned by psychologists to represent the emotion state of texts.
Emotion Expression Assistance: Solving real world problems with sentiment analysis techniques is always attractive, and we have found an
interesting example: helping ESL (English as a Second Language) learners to express their emotions with precise wording. We capture the con-
textual semantics in learners’ writing by identifying emotion events. The goal is to suggest appropriate emotion words according to the context.
Other related NLP research topics I have been working on include opinion mining, medical NLP, and recommendation systems.

Publications

1. Chung-chi Huang and Lun-Wei Ku, “Interest Analysis using 6. Ku, Lun-Wei, Huang, Ting-Hao and Chen, Hsin-Hsi, “Using
PageRank and Social Interaction Content”, Proceedings of the Morphological and Syntactic Structures for Chinese Opinion
6 International Joint Conference of Natural Language Pro- Analysis”, Proceedings of Conference on Empirical Methods
th
cessing (IJCNLP 2013), pages 912-916, October 2013. in Natural Language Processing (EMNLP), pages 1260-1269,
August 2009.
2. Ku, Lun-Wei and Sun, Cheng-Wei, “Detecting Emotion from
Dialogs and Creating Personal Ambient in a Context Aware 7. Ku, Lun-Wei and Chen, Hsin-Hsi, “Mining Opinions from
th
System”, Proceedings of the 15 International Conference on the Web: Beyond Relevance Retrieval”, Journal of American
Human-Computer Interaction (HCI 2013), Lecture Notes in Society for Information Science and Technology, volume 58,
Computer Science, 8028, Springer, pages 128-137, July 2013, number 12, pages 1838-1850, August 2007, Special Issue on
(Also in LNAI, ISI, EI) Mining Web Resources for Enhancing Information Retrieval
3. Ku, Lun-Wei, Sun, Cheng-Wei, and Hsueh, Ya-Hsin, “Dem- 8. Ku, Lun-Wei, Lo, Yong-Shen and Chen, Hsin-Hsi, “Test Col-
onstration of IlluMe: Creating Ambient According to Instant lection Selection and Gold Standard Generation for a Multi-
th
Message Logs”, Proceedings of the System Demonstrations, ply-Annotated Opinion Corpus”, Proceedings of 45 Annual
50 Annual Meeting of Association for Computational Lin- Meeting of Association for Computational Linguistics (ACL),
th
guistics (ACL), pages 97-102, July 2012, pages 89-92, June 2007.
4. Ku, Lun-Wei, Ting-Hao Kenneth Huang, Hsin-Hsi Chen, 9. Ku, Lun-Wei, Ho, Xiu-Wei and Chen, Hsin-Hsi, “Novel Re-
“Predicting Opinion Dependency Relations for Opinion Anal- lationship Discovery Using Opinions Mined from the Web”,
th
ysis”, Proceedings of the 5 International Joint Conference Proceedings of Twenty-First National Conference on Artifi cial
on Natural Language Processing (IJCNLP), pages 345-353, Intelligence (AAAI-06), pages 1357-1362, July 2006,
November 2011.
10. Ku, Lun-Wei, Liang, Yu-Ting and Chen, Hsin-Hsi, “Opinion
5. Ku, Lun-Wei, Ho, Xiu-Wei and Chen, Hsin-Hsi, “Opinion Extraction, Summarization and Tracking in News and Blog
Mining and Relationship Discovery Using CopeOpi Opinion Corpora”, Proceedings of AAAI-2006 Spring Symposium on
Analysis System”, Journal of American Society for Informa- Computational Approaches to Analyzing Weblogs, AAAI
tion Science and Technology, volume 60, number 7, pages Technical Report SS-06-03, pages 100-107, March 2006.
1486-1503, July 2009.

64 研究人員 Research Faculty