SALS-SIG Research Seminar

Text Classification in the Call Center

Abstract: Customer care in technical domains is increasingly based on e-mail communication, allowing for the reproduction of approved solutions. Identifying the customer's problem is often time-consuming, as the problem space changes if new products are launched. This talk describes a new approach to the classification of e-mail requests based on shallow text processing and machine learning techniques. It is implemented within an assistance system for call center agents that was tested in a commercial setting by AOL Germany. The talk describes experiments with different NLP components and ML systems, suggesting that the optimal choice depends on the genre and the language(s) involved. The results of the ICC project led to the foundation of a spin-off company, Xtramind Technologies (http://www.xtramind.com), which pursues the commercial application of customer response management systems. The talk will be preceded by a short overview of intelligent text processing activities currently pursued at DFKI's Language Technology Lab

Bio:

Stephan Busemann is the associate head of DFKI's Language Technology Lab, where he is working as a Principal Researcher and project leader. In 2000, he was appointed DFKI Research Fellow. His areas of expertise are Artificial Intelligence, Computational Linguistics, Language Technology, and Natural Language Generation.

His present research interests are focussed on applied natural language processing: text classification, the application of language generation techniques to summarisation, and the development of solutions based on shallow language generation.

Stephan Busemann studied Computer Science, Artificial Intelligence and Linguistics at the Technical University of Darmstadt and the University of Hamburg. In 1990, he received his PhD from the Computer Science Department of the University of the Saarland with a thesis on NL generation with Generalised Phrase Structure Grammars.

Enquiries: sals@mri.mq.edu.au