Data Warehouse Design For Knowledge Discovery From Healthcare Data

Due to shifting of healthcare data in electronic form, huge volumes of data have been accumulated in organizations related to medical claim processing domain. This paper presents design of a data warehouse implemented as integral part of a rule based expert system. The system is being used for scrubbing of medical claims' data. Approximate star schema has been used for designing the proposed data warehouse. The data warehouse is being used by the production rule mining module and inference engine of the system. Rejected medical claims got corrected in operational database. Payments are received against corrected claims. Production rule mining module needs both versions of a claim, to learn new rules. Therefore data warehouse is required to store all states of medical claims. Moreover, efficiency of the rule based inference engine of the system has increased due to de-normalized format of data in the warehouse.