Interoperable Web-based data mining system by Java distributed object computing

The development of Web-based data mining systems has received a lot of attention in recent years. It plays the key enabling role for competitive businesses in the e-commerce era. A cost-effective and prompt approach for this task is to integrate and coordinate existing data mining applications in a seamless manner. In this paper, we propose a new methodology for developing a Web-based data mining system. This system relies on Java distributed object computing to tackle the issues of interoperability in heterogeneous environments, namely the language, platform, visual object model and data access. The effectiveness of the proposed system is demonstrated by integrating two powerful data mining tools, SOM PAK and Nenet, and an experiment on the Iris data. The methodology can facilitate the collaboration of intelligent components seamlessly in a "plug-N-work" manner, but without re-engineering.