Discovering Keywords for
Search, Covariate Shift and Lifelong
Monday16:00-17:30, 25th, April
TechCode,3F, A, Dinghao Building,Zhongguancun,Beijing
In this talk, I describe three projects that we have been working on.
1. Keyword discovery:
In social media analysis, the user is interested in studying a particular topic. Collecting posts relevant to the topic from a social media data source is a necessary step. Due to the huge size of social media sources (e.g., Twitter, Weibo, and Facebook), the user has to use a set of keywords to search for relevant posts. However, gathering the set of representative topical keywords is a very time-consuming task. Here I discuss our initial work in solving this problem.
2. Covariate shift:
After searching using the keywords, the resulting set of posts can still be quite noisy because many posts containing the keywords may not be relevant. A supervised learning step is needed to filter out those irrelevant posts. Here I discuss a sampling selection bias problem faced by learning, called negative covariate shift, and present an algorithm to deal with it.
3. Lifelong machine learning:
This type of learning retains knowledge learned in the past and uses the knowledge to help future learning. This is in contrast to the current isolated learning paradigm, where a learning algorithm is applied to a given piece of data without considering any related problems and past learned knowledge.
Bing Liu is a professor of Computer Science at the University of Illinois at Chicago. He received his PhD in Artificial Intelligence from the University of Edinburgh. His research interests include sentiment analysis and opinion mining, lifelong machine learning, fake/deceptive opinion detection, data mining, and natural language processing (NLP). He has published extensively in top conferences and journals in these areas. Two of his papers have received 10-year Test-of-Time awards from KDD, the premier conference of data mining and data science.
He also authored three books: two on sentiment analysis and one on Web data mining. Some of his work has been widely reported in the press, including a front-page article in The New York Times. On professional services, he serves as the current Chair of ACM SIGKDD. He has served as program chairs of many leading data mining conferences including KDD, ICDM, CIKM, WSDM, SDM and PAKDD, as associate editors of leading journals such as TKDE, TWEB, DMKD, and as area chairs of numerous NLP, Web research, and data mining conferences. He is a Fellow of ACM,AAAI and IEEE.