Classification of Email Queries by Topic: Approach Based on Hierarchically Structured Subject Domain

Abstract

We describe a Classifier of email queries, which executes text categorization by topic. The specifics of our Classifier is that it allows accurate categorization of short messages containing only a few words. This advantage is achieved by executing morphological and semantic analyses of an incoming text. Specifically, the Classifier provides an efficient information extraction and takes the meaning of words into consideration. By using the hierarchically structured subject domain and classification rules, the Classifier's engine assigns an email query to the most relevant category or categories.