Abstract

Competence is a concept widely recognized as a strategic asset for each organization, especially in areas characterized by intensive use of knowledge, such as scientific communities. Each scientific community aims to produce new knowledge in a particular domain, facilitating knowledge sharing between organizational members with recognized expertises. However, in such a context, scholars are often not aware of other researchers who are working on similar projects, and thus collaboration frequently occurs among a small number of people working in the same group. This work aims to overcome these limitations, proposing an approach oriented to the construction of a scientific community’s competence map that can be used as a mean of highlighting capabilities within the community and suggesting new collaborations between researchers. Traditional approaches in building competence maps are based on tools such as questionnaires or assessment sessions. In recent years, novel approaches based on document content analysis are taking place. However, they do not face up to the problem of defining research competences within a scientific community and surveying them by analysing the scientific production of community members. In this paper we propose an original approach aimed to map scientific competencies, related to research approaches, scientific method phases and field of inquiring, onto scientific community members, by using a weighted multi-hypergraph model. Moreover, by assuming that competences of a researcher can be detected through an analysis of his scientific research papers, we propose a methodological framework to build the map of a many-to-many relationship between researchers and research competencies by means of a collaborative semi-manual mapping process.