Semantics-based cross-domain collaboration recommendation in the life sciences: Preliminary results

In this work we propose a novel approach for semantics-based cross-domain recommendation for research collaboration. First, we construct a large network representing authors, their expertize, current collaborations, and biomedical knowledge in general. We constructed the network from the bibliographic database MEDLINE and from semantic relations extracted from MEDLINE with the SemRep natural language processing system. Then, by using the literature-based discovery paradigm, we recommend novel collaborations, which include not only pairs of authors, but also novel topics for collaboration and an explanation why the collaboration makes sense.