S. Defit (Malaysia) and M.N. Md Sap (Indonesia)
Data Mining, Rough Sets, Association Rules, Useful Knowledge
With the wide applications of computers, database technologies and automated data collection techniques, large amount of data have been continuously collected into databases. It creates great demands for analyzing such data and turning them into useful knowledge. Therefore, it is necessary and interesting to examine how to extract hidden information / knowledge from large amounts of data automatically. Association rules, introduced by Agrawal, Imielinski and Swami, is one of data mining technique to discover interesting rules or relationships among attributes in databases. It has attracted great attention in database research communities in recent years. In this paper, we propose a MML-AR (Mining Multiple Level Association Rules) which integrates rough sets and association rules methods. MML-AR model has been implemented and tested using Jakarta Stock Exchange (JSX) databases. Our study conclude that MML-AR model can improve the performance ability of generated rules.
Important Links:
Go Back
