Association Rules Using Rough Set and Association Rule Methods

摘要

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 or knowledge from large amounts of data automatically and intelligently. In this paper, we propose an MML-AR (Mining Multiple Level Association Rules), which integrates rough set and association rule methods. MML-AR model has been implemented and tested using Jakarta Stock Exchange (JSX) databases. Our study concludes that MML-AR model can improve the performance ability of generated interesting rules.