Toward distributed knowledge discovery on grid systems

Abstract

While massive amounts of data are being collected and stored from not only science fields but also industry and commerce fields, the efficient mining and management of useful information of this data is becoming a challenge and a massive economic need. This led to the development of distributed data mining techniques to deal with huge multi-dimensional datasets distributed among several sites. Besides, to cope with large, graphically distributed, high dimensional, multi-owner, and heterogeneous datasets, Grid platforms are well suited for data storage and they provide an effective computational support for distributed data mining applications. Although Grid platforms allow to share resources distributed in large, heterogeneous environments, there are still many challenges on carrying these distributed data mining techniques on Grid because of lacking efficient distributed data mining systems. In this chapter, we present a new DDM system basing on a Grid/P2P middleware tools to execute new distributed data mining techniques on very large and distributed heterogeneous datasets.