A Framework for Cross-Disciplinary Hypothesis Generation

摘要

The complexity of cross-disciplinary knowledge discovery is twofold: integration of vast amount of information in disparate silos, and dissemination of discovery to stakeholders with different interests. Here we propose a framework that combines Semantic Web technology, graph algorithms, and user profiling to discover and prioritize novel associations among biomedical entities across disciplines. A proof-of-concept system was developed and tested through case studies tailored for three different user groups involved in colorectal cancer (CRC). In this document, we describe in detail the major components of the system and summarize the results of the case studies. The results demonstrate the potential of user profiling and semantic graphs in discovering novel associations that are intellectually engaging to a cross-disciplinary audience.