Colloquium - Lun-Wei Ku

Monday, July 9, 2018 - 11:00am to 12:00pm
HFH 1132
Sentiment and Preference Guided Text Processing in Social Media
Lun-Wei Ku
William Wang


Recent years have witnessed the prosperous development of tools that process natural language text in social media. Many of them, such as web search and instant messaging, have become part of users' life. When using these tools, users leave trails when surfing the web, including both objective and subjective textual information, which we can develop techniques and applications accordingly. The facilitation of the social media environment, like many other environments which can leverage artificial intelligence techniques, could be further enhanced with applications that can well-service and hence engage users, where resonance play a role.  
In this talk I will focus on how we create resonance not only between users but also between users and the developed system by developing the techniques of sentiment analysis, stance classification, and web post recommendation.  I will introduce our work on these research problems from the dataset construction, model design and learning, to system deployment. I will illustrate by examples how recent advances in deep learning can benefit user's emotion and preference learning in a language independent way. As the social media is rife with non-linguistic data, I will also report on our progress and several future directions of learning from heterogeneous features.


Lun-Wei Ku is now an assistant researcher in Institute of Information Science, Academia Sinica, adjunct assistant professor of the national Chiao-Tung university (NCTU), and the secretary-general of Association for Computational Linguistics and Chinese Language Processing. She received her M.S. and Ph.D. degrees from Department of Computer Science and Information Engineering, National Taiwan University. Her research interests include sentiment analysis and opinion mining, natural language processing, information retrieval, and computational linguistics. She has been working on Chinese sentiment analysis since year 2005. She was the co-organizer of NTCIR MOAT Task (Multilingual Opinion Analysis Task, traditional Chinese side) from year 2006 to 2010, and is the organizer of the SocialNLP workshop since 2013. Her international recognition includes CyberLink Technical Elite Fellowship (2007), IBM Ph.D. Fellowship (2008), and ROCLING Doctorial Dissertation Distinction Award (2009). Other professional international activities she involved include: Member-at-Large, Asia Federation of Natural Language Processing; Information Officier (Taiwan Area), ACM SIGHAN; Poster and Demo Chair, AIRS 2018; Financial Chair, IJCNLP 2017; Sentiment Analysis and Opinion Mining, Area Co-Chair, ACL 2017, CCL 2016, ACL-IJCNLP 2015 and EMNLP 2015; NLP Applications, Area Co-Chair, NLPCC 2016; Publication Co-Chair, IJCNLP 2013; Publicity Chair, AIRS 2010.

Everyone welcome!