Automatic Acquisition of a Thesaurus of Interoperability Terms
| Authors: | Velardi Paola, University of Roma, Italy
Navigli Roberto, University of Roma, Italy |
|---|
| Topic: | 5.3 Enterprise Integration and Networking |
|---|
| Session: | Interoperability Research for Networked Enterprises Applications and Software |
|---|
| Keywords: | knowledge acquisition, semantic networks, artificial intelligence, domain analysis, search engines, computer-integrated enterprises, documents |
|---|
Abstract
Enterprise Interoperability is the ability of different organizations to work together and exchange documents, data, services, best practices, etc. A preliminary step towards achieving this goal is the assessment of a common vocabulary of relevant domain concepts, a sort of initial, semi-formal, representation of shared domain knowledge. In this paper we describe an experiment, conducted within the INTEROP Network of Excellence (NoE), aimed at automatically extracting a thesaurus of interoperability terms from the web and a corpus of available documents. Statistical and text mining techniques have been used to extract a glossary of relevant terms and term definitions, as well as a preliminary taxonomic structure of the glossary. The result has been evaluated by a team of partners belonging to the NoE.
