A Rough Set Approach to Compute All Maximal Generalized Rules in Relational Databases

Database stores a huge amount of information in a structured and organized manner and provides many features for machine learning. There are a lot of algorithms to discover different kinds of rules from databases. In this paper, we propose a new method which can compute all maximal generalized rules in relational databases. The method integrates the machine learning paradigm, especially learning-from-examples techniques, with rough-set techniques. An attribute-oriented concept tree ascension technique is first applied in generalization, which substantially reduces the computational complexity of database learning processes. Then the decision matrices are constructed from the generalized relation and the maximal generalized rules with non-necessary constraints can be learned. The authors are members of the Institute for Robotics and Intelligent Systems (IRIS) and wish to acknowledge the support of the Networks of Centres of Excellence Program of the Government of Canada, the Natural ..