A Fuzzy Neural Based Data Classification System

Data mining has emerged to be a very important research area that helps organizations make good use of the tremendous amount of data they have. In data classification tasks, fuzzy systems lack the ability to learn and cannot adjust themselves to a new environment. On the other hand, neural networks can learn, but they are opaque to the user. This paper presents a hybrid system to perform classification tasks. The main work of this paper includes generating a set of weighted fuzzy production rules, mapping it into a min-max neural network; re-deriving the back propagation algorithm for the proposed min-max neural network; and performing data classification. The iris and credit card datasets are used to evaluate the system's accuracy and interpretability. The algorithm has improved the fuzzy classifier.

• Type: Conference
• Year: 2006
• Where: DMIN
• Authors: Luong Trung Tuan, Suet Peng Yong
• Keywords: Classification Tasks, DMIN 2006, DMIN 2007, Min-max Neural Network, Neural Network