Email Analysis and Information Extraction for Enterprise Benefit

keywords: Email, information extraction, trees, graphs, social networks, context, recommendation
In spite of rapid advances in multimedia and interactive technologies, enterprise users prefer to battle with email spam and overload rather than lose the benefits of communicating, collaborating and solving business tasks over email. Many aspects of email have significantly improved over time, but its overall integration with the enterprise environment remained practically the same. In this paper we describe and evaluate a light-weight approach to enterprise email communication analysis and information extraction. We provide several use cases exploiting the extracted information, such as the enrichment of emails with relevant contextual information, social network extraction and its subsequent search, creation of semantic objects as well as the relationship between email analysis and information extraction on one hand, and email protocols and email servers on the other. The proposed approach was partially tested on several small and medium enterprises (SMEs) and seems to be promising for enterprise interoperability and collaboration in SMEs that depend on emails to accomplish their daily business tasks.
reference: Vol. 30, 2011, No. 1, pp. 57–87