Lu Wang      Lu Wang

Ph.D. Candidate
Department of Computer Science
Cornell University
"Natural Language Processing for Understanding Socially-Generated Content”
Monday, February 2, 3:00 PM
STEPS 101

Abstract:   From simple logs to sophisticated proceedings, text has long been integral to knowledge sharing and discovery. Over the past few decades, technology advances have enabled an explosive growth of text data, far outpacing human beings' speed of understanding text. In order to save human's effort and time, there exists a strong need for computer algorithms that improve our ability to find, absorb, and extract information as we need. In this talk, I will present my Ph.D. work on how to use Natural Language Processing (NLP) techniques to (1) address users' information needs, and (2) analyze online social interactions. I will first show an abstract generation system that can automatically produce summaries of the essential output from multi-party meetings. I will also present a socially-informed timeline generation system for complex events. It produces summaries with high continuity that consist of information from both traditional news media and social contexts, for instance, user comments. Finally, I will discuss future directions of my work, which include developing robust NLP algorithms for large-scale data from domains of social importance. In particular, I am interested in domain-specific language understanding and generation techniques for computational social science. I also plan to explore the potential of NLP in health informatics by bridging the gap between sophisticated medical knowledge and ordinary users' information need.

Bio:  Lu Wang is a Ph.D. candidate in Department of Computer Science at Cornell University. She earned bachelor degrees in Machine Intelligence and Economics from Peking University. Lu is interested in developing Natural Language Processing and Machine Learning techniques to help people understand and learn from large-scale text data with inherent noise. She received a best paper nomination award at SIGDIAL 2012. Her paper was also one of the two recipients for Spoken Language Processing Student Travel Award at ICASSP 2014.

© 2014-2016 Computer Science and Engineering, P.C. Rossin College of Engineering & Applied Science, Lehigh University, Bethlehem PA 18015.