Towards a Highly Configurable Data Mining Middleware through Metadata Approach for Access to Unlimited Data Sources

Lai , Ee Hen and Sai , Peck Lee (2007) Towards a Highly Configurable Data Mining Middleware through Metadata Approach for Access to Unlimited Data Sources. In: International Conference on Libraries, Information and Society, ICoLIS 2007, 26-27 June 2007, Petaling Jaya, Malaysia.

[img] PDF
3MY_ Hen Lai Ee_OK.pdf

Download (123kB)
Official URL: http://dspace.fsktm.um.edu.my/handle/1812/284

Abstract

Information stored in a consumer database has become a valuable asset for an organization in today’s information age. It houses vital, hidden information that can be extracted using data mining tools to solve real-world problems in engineering, science, and business. However, due to the rapidly increasing data, data mining tools face substantial challenges in extracting information from data sources such as XML, relational databases, flat files, spreadsheets and so forth. Worst still, these data sources might reside on different locations across different boundaries. Therefore, in order to support new data sources, data mining tools need to provide the flexibility to access the wide range of data sources through straightforward configurations. In this paper, we provide an overview of a proposed data mining middleware known as Java-based Data Mining Middleware which extensively makes use of such configurations through metadata and XML to access unlimited data sources that reside on different locations. Additionally, the proposed middleware also supports a wide range of data mining techniques through trouble-free configurations. Additional data sources and data mining techniques can simply be plugged into the middleware programmatically and non-programmatically.

Item Type: Conference or Workshop Item (Paper)
Subjects: Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science
Divisions: Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
Depositing User: MS SITI NUR ATIKAH MOHAMAD RUSDI
Date Deposited: 23 Jul 2013 08:17
Last Modified: 23 Jul 2013 08:17
URI: http://repository.um.edu.my/id/eprint/436

Actions (login required)

View Item View Item