Mining classification rules using rough sets and neural networks

Classification is an important theme in data mining. Rough sets and neural networks are two common techniques applied to data mining problems. Integrating the advantages of two approaches, this paper presents a hybrid system to extract efficiently classification rules from decision table. Different from those previous works where rough sets were used only for accelerating or simplifying the process of using neural networks for mining knowledge from databases, in our system neural networks are served only as a tool to reduce the decision table and filter its noises while the final knowledge (rule set) is generated from the reduced decision table by rough sets. Therefore, our approach avoids the difficulty of extracting rules from a trained neural network and possesses the robustness which is lacking for rough set based approaches. The effectiveness of our approach was verified by the experiments comparing with traditional rough set and neural network approaches.