Semantic computing technologies have matured to be applicable to many critical domains, such as life sciences and health care. However, the key to their success is the rich domain knowledge which consists of domain concepts and relationships, whose creation and refinement remains a challenge. In this paper, we develop a technique for enriching domain knowledge, focusing on populating the domain relationships. We determine missing relationships between the domain concepts by validating domain knowledge against real world data sources. We evaluate our approach in the healthcare domain using Electronic Medical Record(EMR) data, and demonstrate that semantic techniques can be used to semi-automate labour intensive tasks without sacrificing fidelity of domain knowledge.
Available at: http://works.bepress.com/tk_prasad/79/