Equipe BD
Equipe BD
Laboratoire d'InfoRmatique en Images et Systèmes d'information
UMR 5205 CNRS/INSA de Lyon/Université Claude Bernard Lyon 1/Université Lumière Lyon 2/Ecole Centrale de Lyon

You are here

Extending Email Systems with automatic information extraction technique.

Qui: 
Amin Mesmoudi
Quand: 
Tuesday, September 27, 2011 - 12:30 to 13:00
Où: 
Nautibus

keywords: Email, Information Extraction, Semantic query, Text mining.

Email has become a central method of communication and information exchange in the workplace. However, the use of email is becoming increasingly difficult because the enormous number of emails exchanged every day.
Seek a solution to deal with this problem is the concern of many researchers and developers worldwide. Essentially representing Emails in machine readable format can offer a serious runway to facilitate email related tasks (information search, Meeting planning, etc.)
In this work we investigate the integration of information extraction techniques in Emails. We propose a new technique adapted to emails systems. Our approach is based on two components, entities detection and relations (or dependencies) generation. We extract semantic information (represented in RDF format) from full text emails.
In our new approach, semantic queries are used to exploit Emails using their semantic representation. We study some queries and their evaluation in our system.
We validate our approach by case study, namely, semantic search in emails. This system has been implemented and tested with the Enron benchmark which contains more then 500.000 emails.