Informatică economică (Jan 2009)

An Ontology-centered Approach for Designing an Interactive Competence Management System for IT Companies

  • Stefan TRAUSAN-MATU,
  • Cristina NICULESCU

Journal volume & issue
Vol. 13, no. 4
pp. 159 – 167

Abstract

Read online

The paper presents a generic framework for an intelligent information system of competence management based on ontologies for information technology companies. In a first step it will be applied in an information technology (IT) small enterprise and then its applicability will be verified for other organizations of the same type. The work presented in the paper is performed under the project "CONTO – Ontology-based Competencies Management in Information Technology" funded by the Romanian Ministry of Education and Research, involving two universities, a research institute and an IT private company. A competence management system (CMS), in our vision has to achieve three functions: (a) to support the complete and systematic acquisition of knowledge about the competence of the members of an enterprise; (b) to provide the knowledge about competences and their owners; (c) to apply the available knowledge to serve a purpose. The core of the competence management information system is an ontology that plays the role of the declarative knowledge repository containing the basic concepts (such as: company-job, competence, domain, group, person etc.) and their relationships with other concepts, instances and properties. The Protégé environment was used for the development of this ontology. The structure of the ontology is conceived so that description logics can be used to represent the concept definitions of the application domain in a structured and formally well-understood way. Knowledge acquisition is performed in our approach by enriching the ontology, according to the requirements of the IT company. An advantage of using an ontology-based system is the possibility of the identification of new relations among concepts based on inferences starting from the existing knowledge. The user can choose to query instances of one type of concept. The paper also presents some use-cases.

Keywords