Sigmoid Similarity in Semantic HCP Networks: An Approach for Context Aware Search and Recommendations

Large datasets in silos are available to be harnessed for potential benefits in terms of insights for better recommendations, search and clustering. Healthcare datasets include health care professionals' (HCP) data about their affiliations, publications, participation in conferences, events, clinical trials, specialty and much more. Such datasets in silos are integrated based on the different dimensions such as specialty, participation in various healthcare related events and clinical trials. Integration of healthcare related datasets provide various benefits like better search results, better recommendations of healthcare professionals based on their specialty and needs of patients and better insights to determine key opinion leaders in various therapy areas. In this work, we apply the Sigmoid similarity algorithm to find similar health care professionals based on the HCP Ontology we have developed. Sigmoid similarity is a feature based semantic similarity measure which outperforms the other hierarchy based approaches for computing semantic similarity. This semantic similarity measure enables us to compare and evaluate the semantic similarity amongst various health care professionals to accurately retrieve and recommend healthcare professionals based on the context of search.