Modeling Natural Language Semantics in the Internet Era

By August 18, 2016Events, Seminar

Please register via HKU Event Management System http://goo.gl/ANa4lH

Speaker: Dr. Tin Kam HO (IBM Watson)
Date : Aug 24, 2016 (Wednesday)
Time : 11:00 – 12:30
Venue : Studio 2, 2/F, HKJC BIR, 5 Sassoon Road, Pokfulam.

Abstract

Computational models of learning for audio-visual signal understanding have attracted new attention. In a broader pursuit of artificial intelligence, similar methods are being pushed to another exciting frontier — modeling the semantics of natural languages, concepts, and thought. I will describe our recent attempts on using algorithms to derive and exploit representations of natural language semantics and concept structures. Such representations provide building blocks for larger learning tasks involving sentiments, discoveries, and reasoning.

Research in this area has benefited from many new resources and computational platforms available only through the Internet. I will discuss how it makes use of big data, crowd sourcing, and web services in a unique way. Fierce competitions in the industry are driving the development of applications quickly around any solid progress. There are rich potentials for broader impact from the resources born out of such competitions.

About the speaker

Dr. TIN KAM HO is a Research Staff Member at IBM Watson, where she works in natural language semantics modeling. Formerly she was with Bell Labs, as a Research Scientist and later as the Head of Statistics and Learning Research Department. She pioneered research in multiple classifier systems, random decision forests, and data complexity analysis, and developed many solutions to problems in reading systems, wireless geo-location, smart grid demand forecasting, and optical networks. Her contributions were recognized by several Bell Labs awards, a Young Scientist Award in 1999, and the Pierre Devijver Award for Statistical Pattern Recognition in 2008. She is an elected Fellow of the International Association for Pattern Recognition and the IEEE. She served as Editor and Associate Editor for several journals, and as Editor-in-chief of Pattern Recognition Letters in 2004-2010. She received a Ph.D. in Computer Science from SUNY at Buffalo in 1992.

 

Please register via HKU Event Management System http://goo.gl/ANa4lH

Enquiries: Please contact Mr. Rickey YAU at csrp@hku.hk or 28315232