Job-Candidate Matching using ESCO Ontology

Abstract

 Skills management is one of the key factors to address the increasing competitiveness among different companies. Suitable knowledge representation and approach for matching skills and competences in job vacancies and candidate profiles can support human resources management automation through suitable matching and ranking services. This paper presents an approach for matchmaking between skills demand and supply through skill profiles enrichment and matching supply and demand profiles over multiple criteria. This work builds upon methods for profile modeling, information enrichment and multi-criteria matching. The main contribution of this work is a methodology for harmonization and enrichment of heterogeneous profile models and skill set description by making use of the standard ESCO ontology. Secondly, an algorithm is proposed for similarity matching across multi-criteria for discovering set of profiles that best fits the job description criteria. A prototype web-based system has been developed to implement the proposed approach and deployed online. The system has been tested with real IT jobs related dataset and validated against relevance scores provided by human experts. Experimental results show consistent correspondence between the similarity ranking scores produced by the system and scores provided by the human users.