Structural analysis of chat messages for topic detection

Purpose This paper studies the characteristics of chat messages from analyzing a collection of 33,121 sample messages gathered from 1700 sessions of conversations of 72 pairs of MSN Messenger users over 4-month duration from June to September of 2005. The primary objective of chat message characterization is to understand the properties of chat messages for effective message analysis such as message topic detection. Methodology/Approach From the study on chat message characteristics, an indicative term-based categorization approach for chat topic detection is proposed. In the proposed approach, different techniques such as sessionalization of chat messages and extraction of features from icon texts and URLs are incorporated for message pre-processing. And Näıve Bayes, Associative Classification, and Support Vector Machine are employed as classifiers for categorizing topics from chat sessions. Findings Indicative term-based approach is superior than the traditional document frequency based approach for feature selection in chat topic categorization. Originality/Value This paper studies the characteristics of chat messages and proposes an indicative term-based categorization approach for chat topic detection. The proposed approach has been incorporated into an instant message analysis system for both online and offline chat topic detection. Preprint submitted to Online Information Review 2 May 2006