The design and implementation of OGSA-DAI-DM

Abstract

The transition from centralized data mining process as they are in traditional tools to Grid-based data mining services that can coordinate with each other is crucial to mine useful knowledge from distributed data resources. So far, data mining process has not been integrated with the current Grid architecture, and how to extract knowledge from distributed data resources managed by Grid platform is still a challenge for many data mining applications. This paper presents a data mining middleware that extends OGSA-DAI to extract knowledge from distributed data sources, and we call it OGSA-DAI-DM, which aims to be one of reference implementations for the coming WS-DAI-DM interface specification. Several OGSA-DAI activities and workflows for handling association rules and sequential patterns mining are implemented. © 2011 Springer-Verlag.