Acquisition of a Key Concept Dictionary for Classifying E-mails

S. Sakurai, A. Suyama, and K. Fume (Japan)

Keywords

Text Mining, Fuzzy Inductive Learning, Email Classification, Customer Center

Abstract

This paper proposes a method to classify e-mails collected in a customer center by using text mining. The method uses two kinds of domain-dependent knowledge. One is a key concept dictionary provided by a human expert. The other is a concept relation dictionary in the form of a fuzzy de cision tree automatically acquired from training examples. The method inputs the subject and the body of an e-mail and decides a text class for the e-mail. This paper applies the method to three kinds of classification tasks: a prod uct classification task, a contents classification task, and an address classification task. The results of numerical exper iments indicate that acquired concept relation dictionaries give highly precise ratios.

Important Links:



Go Back