Title :
Strategies for parallel data mining
Author :
Skillicorn, David
Author_Institution :
Dept. of Comput. & Inf. Sci., Queen´´s Univ., Kingston, Ont., Canada
Abstract :
This article presents a set of cost measures that can be applied to parallel algorithms to predict their computation, data access and communication performance. These measures make it possible to compare different parallel implementation strategies for data mining techniques without benchmarking each one
Keywords :
data mining; parallel algorithms; software performance evaluation; benchmarking; communication performance; computation; cost measures; data access; parallel algorithms; parallel data mining; parallel implementation strategies; Concurrent computing; Costs; Data mining; Marketing and sales; Testing;
Journal_Title :
Concurrency, IEEE
DOI :
10.1109/4434.806976
