关键词: 增量学习, 模糊关联分类, eVQ聚类, UWEP, 高斯隶属度函数

Abstract: Abstract: To improve the efficiency of building the FAC (Fuzzy Associative Classifier) on the dynamic data set, an incremental fuzzy associative classification method based on eVQ(evolving Vector Quantization) cluster algorithm is proposed. Firstly, eVQ cluster algorithm was adopted to incrementally update the parameters of Gauss membership functions of quantitative attributes. Secondly, UWEP (Update With Early Pruning) algorithm was extended to incrementally mine fuzzy frequent itemsets. Finally, FCORR (Fuzzy Correlation) of fuzzy associative classification rule (FACR) and the length of antecedent of FACR were regarded as measure to prune and update fuzzy associative classification rule base. The experiment results on four UCI benchmark data sets show that compared with the method of the batch building the FAC, the proposed method can reduce the time of training the FAC in the premise of not decreasing the accuracy and interpretability. The Gauss membership function updating method based on eVQ cluster algorithm contributes to improve the classification accuracy of the FAC on the dynamic data set.

Key words: Keywords: Incremental learning, Fuzzy associative classification, eVQ cluster, UWEP, Gauss membership function