Weighted doubly regularized support vector machine and its application to microarray classification with noise

详细信息    查看全文

• 作者：Juntao Li^a ; ^{juntaolimail@126.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Yadi Wang^a ; ^{wangyadime@126.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Yimin Cao^a ; ^{cao_yimin@126.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Cunshuan Xu^b ; ^{cellkeylab@126.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae
• 关键词：Support vector machine ; Double weights ; Gene selection ; Microarray classification ; Solution path algorithm
• 刊名：Neurocomputing
• 出版年：2016
• 出版时间：15 January 2016
• 年：2016
• 卷：173
• 期：part_P3
• 页码：595-605
• 全文大小：522 K

文摘

A weighted doubly regularized support vector machine was proposed and its solution path algorithm was developed. By using both the distance information between classes and within each class, a double-weighted mechanism was presented, based on which the weighted doubly regularized support vector machine was proposed. A staircase function between two model parameters along the solution path direction was proposed, the multiple parameter selection problem was transformed into single parameter problem and the corresponding solution path algorithm was developed. The proposed support vector machine can adaptively identify the important genes in groups, thus encouraging an adaptive grouping effect. The experiment results on leukaemia, colon cancer, lung cancer data sets demonstrated that the proposed method can effectively select genes and reduce the influence of noise.