LICT center Distinguished Lecture Program - Lecture on Text Mining
Lecture on 'Text Mining - Creating Semantics in the real world', by Prof. Stefan Wrobel
LICT (Leuven ICT center) kindly invites you to this talk that will take place on April 26th, 2010 at 16:30.
The session is open for all interested parties.
Participation is free of charge, but advance registration is asked for.
ABOUT THE EVENT:
Date: April 26th, 2010
Location: K.U.Leuven, Auditorium 200 N.00.04, Celestijnenlaan 200, Leuven, Belgium
Among the wealth of data available today, the large majority is not in a structured format, but consists of texts and multimedia documents. Library archives, company hard disks and the world wide web offer billions and billions of documents, but string-based search lacks the necessary semantics to find and identify the valuable among the irrelevant. In recent years, research in text mining has shown that by relying on robust and simple linguistic pre-processing combined with powerful Machine Learning techniques that exploit the statistical properties of large document collections, it is possible to fill at least part of the semantic void by classifying documents, structuring and analyzing them for retrieval, or monitoring document collections. In this talk, following a brief introduction into the basic challenges and approaches in text mining, we will focus on project examples from our own work that demonstrate that text mining has arrived in the real world. We will show our text mining can help identify fake merchandise and phishing emails, how it can help structure large document collections from leading newspapers and cultural institutions, and how it can lead to web portals offering simple but effective structuring. We will give an outlook to what might be possible in monitoring the blogosphere and show perspectives beyond text.
Bio of speaker:
Stefan Wrobel holds a joint position as Managing Director of the Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS and Chair of Computer Science/Intelligent Analysis and Information Systems at University of Bonn.
This lecture fits in the Distinguished Lecture Program of LICT.
Participation is free of charge, but advance registration is asked for