Lehigh University - CSE: Hurst

Dr. Matthew Hurst

"Mining Opinion from the Web"

Thursday, November 18, 12 PM

Packard Lab, Room 101

Abstract: The recent popularity of online publication, primarily in the form of weblogs, has focused a number of applications on the various genres of online consumer data. In this talk, I will present an end-to-end system which gathers online consumer data, applies a number of advanced analysis methods and produces insights into the opinions and attitudes of the web-public. This technology is currently used in industrial applications, servicing a number of corporate and government users, including market analysts and product managers.

The system elements include: focused crawling, wrapper induction, document analysis, data driven phrase mining, document classification, topic analysis, sentiment analysis and other text and data mining features.

Bio: Matthew Hurst received his PhD from Edinburgh University. After a European Commission Fellowship spent in Japan at the Electrotechnical Laboratory in Tsukuba, he was a visiting scientist at IBM Research Japan where he worked with the text mining group. He has continued research into document analysis and text mining in industrial settings at WhizBang!Labs, and currently at Intelliseek, Inc. where he is a Senior Research Scientist and co-founder of BlogPulse.

		Dr. Matthew Hurst "Mining Opinion from the Web" Thursday, November 18, 12 PM Packard Lab, Room 101 Abstract: The recent popularity of online publication, primarily in the form of weblogs, has focused a number of applications on the various genres of online consumer data. In this talk, I will present an end-to-end system which gathers online consumer data, applies a number of advanced analysis methods and produces insights into the opinions and attitudes of the web-public. This technology is currently used in industrial applications, servicing a number of corporate and government users, including market analysts and product managers. The system elements include: focused crawling, wrapper induction, document analysis, data driven phrase mining, document classification, topic analysis, sentiment analysis and other text and data mining features. Bio: Matthew Hurst received his PhD from Edinburgh University. After a European Commission Fellowship spent in Japan at the Electrotechnical Laboratory in Tsukuba, he was a visiting scientist at IBM Research Japan where he worked with the text mining group. He has continued research into document analysis and text mining in industrial settings at WhizBang!Labs, and currently at Intelliseek, Inc. where he is a Senior Research Scientist and co-founder of BlogPulse.