166 Int. J. Metadata, Semantics and Ontologies, Vol. 6, Nos. 3/4, 2011 Copyright © 2011 Inderscience Enterprises Ltd. Using ontology for resume annotation Wahiba Ben Abdessalem Karaa* and Nouha Mhimdi High Institute of Management, 41, Rue de la Liberté, Cité Bouchoucha, 2000 Le Bardo, Tunis, Tunisia E-mail: wahiba.abdessalem@isg.rnu.tn E-mail: nouhamhimdi@yahoo.fr *Corresponding author Abstract: Employers collect a large number of resumes from job portals or from their company’s own website. These documents are used for an automated selection of candidates satisfying the requirements and therefore reducing recruitment costs. Various approaches for process documents have already been developed for recruitment. In this paper, we present an approach based on semantic annotation of resumes for an e-recruitment process. The most important task consists of modelling the semantic content of these documents using ontology. The ontology is built taking into account the most significant components of resumes inspired from the structure of EUROPASS CV. This ontology is thereafter used to annotate automatically the resumes. Keywords: annotation; e-recruitment; ontology; semantic web; text analysis; resume; CV. Reference to this paper should be made as follows: Ben Abdessalem Karaa, W. and Mhimdi, N. (2011) ‘Using ontology for resume annotation’, Progress in Computational Fluid Dynamics, Vol. 6, Nos. 3/4, pp.166–174. Biographical notes: Wahiba Ben Abdessalem Karaa is an Assistant Professor in the Department of Computer and Information Science at the University of Tunis, High Institute of Management. She received her Master’s Degree in 1992 from Paris III, New Sorbonne, France, and her PhD from Paris 7, Jussieu, France, in 1997. Her research interests include modelling information systems, Natural Language Processing, document annotation, information retrieval, text mining, knowledge extraction, semantic web, etc. She is a member of the program committee of several international conferences: ICCA’2010, ICCA’2011, RFIW 2011, and a member of the Editorial Board of the International Journal of Managing Information Technology (IJMIT). Nouha Mhimdi is a PhD Candidate at the University of Tunis in the High Institute of Management. She received her Master’s Degree in Computer Science in 2010 from the High Institute of Management. Her research interests include Natural Language Processing, text mining, semantic annotation, and ontologies. 1 Introduction Employers often receive a large number of resumes for an open position. The costs of classically and manually selecting appropriate candidates have amplified, and employers are searching for tools to automate the candidate’s selection. Online recruitment processes can be efficient using semantic web technologies (Bizer et al., 2005). Indeed, using semantic web technologies in the recruitment domain could considerably decrease the costs for employers in terms of publishing job offers and selecting candidates (Morin et al., 2004). In this paper, we investigate how semantic web technologies can be used to support the recruitment process. In particular, we describe the design and implementation of an ontology which defines the semantic content of the resumes. We propose in addition an annotation approach of resumes based on this ontology. We also present the results of initial experiments testing the performance of this prototype implementation. The rest of the paper is organised as follows: Section 2 gives a brief overview of related works of the e-recruitment field. We describe our approach for ontology creation, and the global architecture of the annotation system in Section 3. In Section 4, an evaluation of our approach is shown, and then a conclusion and perspectives for improving this work are given at the end of this paper. 2 Related works In the field of e-recruitment, the information contained in the resume of a candidate who applies for a job is specific, and requires semantic and automatic processing. In this context, several approaches conducted under the recruitment (e-recruitment), take into account the semantics of used documents (e.g., resume). Some of them index resumes using semantic techniques that consist mainly of the association of the most important elements in a document to concepts existing in the resume. This approach is based