CRNN: Integrating Classification Rule and Neural Network

Association classification has been an important type of the rule based classification. A variety of approaches have been proposed to build a classifier based on classification rules. In the prediction stage of the extant approaches, most of the existing association classifiers use the ensemble quality measurement of each rule in a subset rules to predict the class label of the new data. This method still suffers the following two problems. The classification rules are used individually that the coupling relations between rules are ignored in the prediction. However, in real-word rules set, the rules are often inter related together and many rules are usually satisfied partially when a new data object comes. Furthermore, the classification rule based prediction model lacks a general expression of the decision methodology. This paper proposes a classification method that integrating classification rule and neural network (CRNN for short), which gives a general form of the rule based decision methodology by rules network. In comparison with the extant rule based classifiers, such as C4.5, CBA, CMAR and CPAR, our approach has two advantages. First, CRNN takes into account of the coupling relations between rules from the training data in the prediction step. Second, CRNN obtains higher performance on the structure and parameter learning automatically than traditional neural network. CRNN uses the linear computing algorithm in neural network instead of the costly iterative learning algorithm. Two ways of the classification rule set generation are conducted in this paper for the CRNN evaluation and CRNN achieves the satisfied performance.