Interactive, Topic-Based, Visual Text Analytics
Time：16:00, May 9, 2014
Venue：Meeting Room 202, Office Building,
Software Campus
Speaker：Dr. Shixia Liu, Microsoft Research Asia
Host：Dr. Lu Lin
Abstract:
Businesses use text documents to communicate with their shareholders, share knowledge within the enterprise, coordinate activities among employees, and track business processes. As a result, analyzing text documents has become increasingly an important part of decision making in large corporations and small businesses. In this talk, I’ll present some of our research results and use them to exemplify how we solve the challenges in real-world applications. Specially, I’d like to introduce 1) An interactive visual analysis tool called TextFlow to help users analyze how and why the correlated topics change over time; 2) TextPioneer for identifying the topics in a text corpus which lead the similar topics in other corpora; 3) TopicPanorama for developing a full picture of relevant topics discussed in multiple sources such as news, blogs, or micro-blogs. In the aforementioned work, we aim to demonstrate how visual text analytics techniques can help people analyze large collections of text and make decisions.
Bio:
Dr. Shixia Liu is a lead researcher in the Internet Graphics Group at Microsoft Research Asia. She received a BSC degree and a MSC degree in Computing Mathematics from Harbin Institute of Technology, a Ph.D. in Computer Aided Design and Computer Graphics from Tsinghua University. Before she joined MSRA, She worked as a research staff member and research manager at IBM China Research Lab. Her research interests include interactive, visual text analytics and interactive, visual social analytics, and visual search log analysis. She is the program co-chair of VINCI'2012 and IEEE VIS 2014 Meetup. She is also the guest editors of ACM Transactions on Intelligent Systems and Technology, and Tsinghua Science and Technology. She is in the program committee of InfoVis, VAST, KDD, PacificVis, ACM Multimedia, SDM, IUI, VINCI, IVAPP, and PAKDD.
