An Improved Breast Cancer Diagnosis System Utilizing BPSO-based Feature Selection

摘要：

In this paper a novel accurate system for breast cancer diagnosis is introduced. The proposed system is based on a high performance feature selection by binary particle swarm optimization (BPSO). The diagnostic system contains 4 steps: dataset normalization, feature selection, support vector machine (SVM) classifier training and fusion of 5 top classifiers using majority rule. As an effective part of this work, feature selection is performed utilizing BPSO to reduce feature vector dimension. Merit of the system is successfully certified on WDBC dataset leading to recognition rate of％100.The proposed system clearly outperforms previous works in both respects of accuracy and number of required features. Also using low dimension feature vectors would increase speed of the proposed system.

关键词： Breast cancer diagnosis Feature selection BPSO SVM FNA test

作者: M.Alipoor S.E.Abtahi J.Haddadnia

作者单位: Engineering Department of Tarbiat Moallem University of Sabzevar,Sabzevar,Iran

会议类型: 国际会议

会议名称: 2009年中国控制与决策会议(2009 Chinese Control and Decision Conference)

会议地点: 广西桂林

会议语种:英文

页码: 4962-4967

在线出版日期: 2009-06-17（万方平台首次上网日期，不代表论文的发表时间）