The Research and Implementation of Grid Based Data Mining Architecture

摘要

Nowadays E-Business and E-Science are generating plenty of datasets. These datasets are heterogeneous and geographically distributed. There are major challenges involved in the efficient extracting useful knowledge from the datasets. This paper proposes a Grid based data mining architecture for Grid based Urban Public Transport Decision Support System (GUPTDSS). It discusses three main topics: process of parallel algorithm; deployment, invoking and scheduling of Grid based data mining service; data sources distribution scenarios and data access. To evaluate the efficiency of the proposed system, an example of traffic flow classification is presented.