Design and Implementation of WSRF-Compliant Grid Services for Mining Fuzzy Association Rules


Department of Computer Science and Engineering,Shiraz University


Data mining is a widely used approach for the transformation of large amounts of data
to useful patterns and knowledge. Fuzzy association rules mining is a data mining technique which tries
to nd association rules without the e ect of sharp boundary problems when data contains continuous
and categorical attributes. Grid data mining is a new concept, which allows the data mining process
to be deployed and used in a data grid environment where data and service resources are geographically
distributed. In this paper, a grid service for mining fuzzy association rules is developed. The service
is implemented based on recently proposed Data Mining Grid Architecture (DMGA) and uses the Web
Service Resource Framework (WSRF). Experimental evaluations, after implementing and deploying the
service, show the e ectiveness and acceptable performance of the proposed grid service. Additionally, in
this study, a new algorithm, namely FFDM, is developed to mine fuzzy association rules without raw data
exchange, using the distributed storage of data grid environments. Empirical evaluation of FFDM reveals
the scalability and eciency of the proposed method, in addition to the advantages of minimum messaging
and providing privacy of data.