A neural-networks associative classification method for association rule mining

Abstract

Association rule theory has been used to analyze relationships between data in a large database. It has been applied in various fields including market analysis, customer habit study, detection, classification, clustering and so on. However, it consumes a great amount of time. In the Apriori approach, all frequent patterns are found and then generate association rules from above frequent patterns. However, this approach generates a huge time-intensive query called "iceberg query". Various researches have been done under the Apriori-like approach to improve performance of the frequent pattern mining task but the results were not as much as expected due to many scans on the dataset. This research aims to propose a flexible way of mining frequent patterns by extending the idea of the Associative Classification method. For better performance, the Neural-Network Associative Classification (NAC) method is proposed here to be one of the approaches for building accurate and efficient classifiers. In this paper, the Neural Networks Associative Classification method is used in order to improve its accuracy. The structure of the network reflects the knowledge uncovered in the previous discovery phase. The trained network is then used to classify unseen data. The accuracy rates obtained from the datasets show promising results.