Promoting Research Collaboration Based on Data Mining Techniques in Library Information Systems

Research  collaboration  connects  distributed  knowledge  and  competencies  into  new  ideas  and research institutes and has been the subject of many research projects. We argue that academic libraries, including libraries of universities and research institutes, hold a wealth of information regarding patrons' research interests hidden in their data bases. Mining these databases can provide better understanding of researchers' needs and interests. This paper has two main contributions. Firstly, it proposes a new methodology based on data mining techniques in library information systems to uncover patrons' research interests in order to facilitate research collaboration including interdisciplinary research. The proposed methodology, studies data mining techniques in a library information system  as  a  case  study  and  makes  advantage of clustering  algorithms  to  cluster researchers based on their library usage which is interpreted as their research interests. The second contribution of this paper is that, it presented a knowledge map as a visual representation of usage trends of an academic library to portray virtual interest groups based on item use information. The result of  this study can support managers and decision makers for strategic decision making regarding future research directions and collaborations. The outcome of the case study confirms our hypotheses by revealing clusters of library users with similar research interests validated by their academic backgrounds.