TY - GEN
T1 - Decomposition methodology for classification tasks
AU - Rokach, Lior
AU - Mainon, Oded
PY - 2005/12/1
Y1 - 2005/12/1
N2 - The idea of decomposition methodology is to break down a complex data mining task into several smaller, less complex and more manageable, sub-tasks that are solvable by using existing tools, then joining their solutions together in order to solve the original problem. In this paper we provide an overview of decomposition methods in classification tasks with emphasis on elementary decomposition methods. We present the main properties that characterize various decomposition frameworks and the advantages of using these framework. Finally we discuss the uniqueness of decomposition methodology as opposed to other closely related fields, such as ensemble methods and distributed data mining.
AB - The idea of decomposition methodology is to break down a complex data mining task into several smaller, less complex and more manageable, sub-tasks that are solvable by using existing tools, then joining their solutions together in order to solve the original problem. In this paper we provide an overview of decomposition methods in classification tasks with emphasis on elementary decomposition methods. We present the main properties that characterize various decomposition frameworks and the advantages of using these framework. Finally we discuss the uniqueness of decomposition methodology as opposed to other closely related fields, such as ensemble methods and distributed data mining.
UR - http://www.scopus.com/inward/record.url?scp=33845292574&partnerID=8YFLogxK
U2 - 10.1109/GRC.2005.1547369
DO - 10.1109/GRC.2005.1547369
M3 - Conference contribution
AN - SCOPUS:33845292574
SN - 0780390172
SN - 9780780390171
T3 - 2005 IEEE International Conference on Granular Computing
SP - 636
EP - 641
BT - 2005 IEEE International Conference on Granular Computing
T2 - 2005 IEEE International Conference on Granular Computing
Y2 - 25 July 2005 through 27 July 2005
ER -