A Framework For Knowledge Discovery in Multivalued Tables Using Rough Set Approach

摘要

In data mining area, researchers have proposed methods to discover decision rules that are based on rough set approach. Most of these methods do not have ability to handle multivalued information systems, though this type of information systems appears frequently in practical problems. In this paper, we propose an approach based on rough set to handle multivalued information systems. With the model designed from this approach, we propose two algorithms to discover minimal decision rules from multivalued information systems. In addition, we make a comparison between this method and other classical ones.