Rules mining from multi-layered neural networks

Abstract

Data mining (DM) is a process of non-trivial extraction of implicit, previously unknown and potentiality useful information from a large volume of data. The mined information is also referred as knowledge of the form rules, constraints and regularities. Rule mining is one of vital tasks in DM since rules provide a concise statement of potentially important information that is easily understood by end users. Researchers have been using many techniques such as statistical, AI, decision tree, database, cognitive, etc., for rule mining. Rule mining using neural networks (NNs) is a challenging job as there is no straight way to translate NN weights to rules. However, NNs have potential to be used in rule mining since they have been found to be a powerful tool to efficiently model data and modelling data is also an essential part of rule mining. Considering these powerful features of neural networks, our paper will propose a novel rule mining algorithms from multi-layered neural networks. The proposed rule mining methods clusters attributes rather than examples of data and from clusters rule mining becomes fast and manageable. Moreover, two constraints support and confidence give the controls over the mining result of the proposed method.