A Hybrid Method to Discover and Rank Cross-Disciplinary Associations

As basic science (ldquobenchrdquo) and medical practice (ldquobedsiderdquo) continue their exponential growth in complexity and scope, the need for finding hidden connections and translating knowledge across disciplines becomes inevitable. The proposed method combines semantic Web technology, graph algorithms, and user profiling to discover and prioritize novel cross-disciplinary associations based on each user's interest. A proof-of-concept system was developed and tested through a set of use cases. In each use case, novel associations are suggested and ranked by the system for individual user. We demonstrated the potential of the proposed method in facilitating hypothesis generation and knowledge translation across disciplines.