Decomposition Methodology for Knowledge Discovery and Data Mining资料.pdfVIP

Chapter 46 DECOMPOSITION METHODOLOGY FOR KNOWLEDGE DISCOVERY AND DATA MINING Oded Maimon Department ofIndustrial Engineering Tel-Aviv University maimon@eng.tau.ac.il Lior Rokach Department ofIndustrial Engineering Tel-Aviv Universiry liorr@eng.tau.ac.il Abstract The idea of decompositionmethodology is to break down a complex Data Min- ing task into several smaller, less complex and more manageable, sub-tasks that are solvable by using existing tools, thenjoining their solutions together in or- der to solve the original problem. In this chapter we provide an overview of decomposition methods in classification tasks with emphasis on elementary de- composition methods. We present the main properties that characterize various decomposition frameworks and the advantages of using these framework. Fi- nally we discuss the uniqueness of decomposition methodology as opposed to other closely related fields, such as ensemble methods and distributed data min- ing. Keywords: Decomposition, Mixture-of-Experts, Elementary Decomposition Methodology, Function Decomposition, Distributed Data Mining, Parallel Data Mining 1 Introduction One of the explicit challenges in Data Mining is to developmethods that will be feasible for complicated real-world problems. In many disciplines,when a problem becomes more complex, there is a natural tendency to try to break it down into smaller, distinct but connected pieces. The concept of breaking 982 DATA MINING AND KNOWLEDGE DISCOVERY HANDBOOK down a system into smaller pieces is generally referred to as decomposition. The purpose of decomposition methodology is to break down a complex prob- lem into smaller, less complex and more manageable, sub-problems that are solvable by using existing tools, th