Collective Data Mining: a New Perspective toward Distributed Data Mining Advances in Distributed Data Mining Book

This paper introduces the collective data mining (CDM), a new approach toward distributed data mining (DDM) from heterogeneous sites. It points out that naive approaches to distributed data analysis in a heterogeneous environment may face ambiguous situation and may lead to incorrect global data model. It also observes that any function can be expressed in a distributed fashion using a set of appropriate basis functions and orthonormal basis functions can be eeectively used for developing a general framework for DDM that guarantees correct local analysis, resulting in desired global data model using minimal data communication. The paper develops the foundation of CDM, discusses decision tree learning and polynomial regression in CDM for discrete and continuous variables, and describes the BODHI, a CDM based experimental system.