Pattern Recognition Based on Weighted and Supervised ART2

摘要：

It is crucial for TCM(Traditional Chinese Medicine)post-hepatitis cirrhosis diagnosis to accurately identifythe syndrome.Meanwhile,the selection of featureswhich are relevant to a certain TCM post-hepatitiscirrhosis syndrome not only improves the performanceof the classifiers,but also provides well measure fortreatment.Therefore,in this paper,we analyze theclassical ART2(Adaptive Resonance Theory 2)neuralnetwork,such as the problem of pattern drifting andthe same phase data with different amplitudes.Basedon this,here,a novel network named SWART2 isproposed by taking dispersion testing and centroidcomputation learning,and introducing the weightedand supervised mechanism,which aims at improvingART2s ability of classification greatly for post-hepatitis cirrhosis diagnosis.Experimental results inthis paper showed that the new SWART2 performedbetter than classical ART2.

作者: Chu Na Ma Lizhuang

作者单位: Dept.of Computer Science & Engineering,Shanghai Jiao Tong University,Shanghai,China

会议类型: 国际会议

会议名称: 2008 3rd International Conference on Intelligent System and Knowledge Engineering(第三届智能系统与知识工程国际会议)(ISKE 2008)

会议地点: 厦门

会议语种:英文

页码: 98-102

在线出版日期: 2008-11-17（万方平台首次上网日期，不代表论文的发表时间）