Clustered Collaborative Filtering Approach for Distributed Data Mining on Electronic Health Records

Distributed Data Mining (DDM) has become one of the promising areas of Data Mining. DDM techniques include classifier approach and agent-approach. Classifier approach plays a vital role in mining distributed data, having homogeneous and heterogeneous approaches depend on data sites. Homogeneous classifier approach involves ensemble learning, distributed association rule mining, meta-learning and knowledge probing. Heterogeneous classifier approach involves collective principle component analysis, distributed clustering, collective decision tree and collective bayesian learning model. In this paper, classifier approach for DDM is summarized and an architectural model based on clustered-collaborative filtering for Electronic Health Records (EHR) is proposed.