Organizing open archives via lightweight ontologies to facilitate the use of heterogeneous collections

Purpose – This paper seeks to focus on the problems of integrating information from open, distributed scholarly collections, and on the opportunities these collections represent for research communities in developing countries. The paper aims to introduce OntOAIr, a semi‐automatic method for constructing lightweight ontologies of documents in repositories such as those provided by the Open Archives Initiative (OAI). Design/methodology/approach – OntOAIr uses simplified document representations, a clustering algorithm, and ontological engineering techniques. Findings – The paper presents experimental results of the potential positive impact of ontologies and specifically of OntOAIr on the use of collections provided by OAI. Research limitations/implications – By applying OntOAIr, scholars who frequently spend many hours organizing OAI information spaces will obtain support that will allow them to speed up the entire research cycle and, expectedly, participate more fully in global research communities. Originality/value – The proposed method allows human and software agents to organize and retrieve groups of documents from multiple collections. Applications of OntOAIr include enhanced document retrieval. In this paper, the authors focus particularly on document retrieval applications.