Recommending Scientific Collaboration Based on Topical, Authors and Venues Similarities

Scientific communities can be considered as a particular kind of practice communities. They consist of various interdisciplinary groups that are often geographically distributed. A common area of research interests can bring them together. Scientific production growth has been increasing the difficulties for identifying the most relevant article to cite or to find an appropriate conference or journal to submit an article for publication or a potential author to collaborator with. In this paper, we propose an algorithm that analyzes the article manuscript along with the cited articles to determine the most suitable venue and a potential author to collaborator with. Our algorithm exams the topical similarity, the authors' similarities, and the venues' similarities for a given query to estimate the collaborative confident score that represents the anticipation of a prospective relationship between the researchers among a scientific community. The effectiveness of our approach is evaluated based on a real academic dataset.