Keeping Your Threads Untangled: An Intelligent System for Semi-automatically Organising Corporate Messages by Content

This paper describes an approach, based on a system, called Threads, which is already in commercial use, to automatically synchronise sets of digital messages (e.g. e-mail and digitised phone calls) with particular projects, in which context the messages were created. Based on the authors' experience that it is generally easier to browse projects of interest than to search masses of messages, authorised users can extract information from far more messages, far quicker than they might otherwise do by trying to guess suitable search terms. The novelty described in this paper is a method for organising such messages automatically and intelligently according to statistics relating to less common words (called "keywords") which they contain. Initial experiments using this method are described, using both e-mail data from the parent company, and on the publicly available ENRON dataset of e-mails and phone calls. These preliminary results are interesting, but suggest the method as currently implemented is only suitable for semi-automatic classification, but not yet for fully automated allocations to projects.