Fusion of Decision Tree and Gaussian Mixture Models for Heterogeneous Data Sets

Current data mining techniques have been developed with great success on homogeneous data. However, few techniques exist for heterogeneous data without further manipulation or consideration of dependencies among the different types of attributes. This paper presents a fusion of C4.5 Decision Tree and Gaussian Mixture Model (GMM) techniques for mixed-attribute data sets. The proposed fusion technique is used to detect anomalies in computer network data. Evaluation experiments were performed on the popular KDDCup 1999 data set using C4.5 Decision Tree, GMM and fusions of C4.5 and GMM. Experimental results showed a better performance for the proposed fusion technique compared to the individual techniques.