Mutual information-based SVM-RFE for diagnostic classification of digitized mammograms Export

@article{citeulike:5360516, abstract = {Computer aided diagnosis (CADx) systems for digitized mammograms solve the problem of classification between benign and malignant tissues while studies have shown that using only a subset of features generated from the mammograms can yield higher classification accuracy. To this end, we propose a mutual information-based Support Vector Machine Recursive Feature Elimination (SVM-RFE) as the classification method with feature selection in this paper. We have conducted extensive experiments on publicly available mammographic data and the obtained results indicate that the proposed method outperforms other SVM and SVM-RFE-based methods.}, author = {Yoon, Sejong and Kim, Saejoon}, citeulike-article-id = {5360516}, citeulike-linkout-0 = {http://dx.doi.org/10.1016/j.patrec.2009.06.012}, citeulike-linkout-1 = {http://linkinghub.elsevier.com/retrieve/pii/S0167865509001639}, day = {01}, doi = {10.1016/j.patrec.2009.06.012}, issn = {01678655}, journal = {Pattern Recognition Letters}, keywords = {feature\_selection}, month = {December}, number = {16}, pages = {1489--1495}, posted-at = {2009-10-31 10:57:02}, priority = {2}, title = {Mutual information-based SVM-RFE for diagnostic classification of digitized mammograms}, url = {http://dx.doi.org/10.1016/j.patrec.2009.06.012}, volume = {30}, year = {2009} }

TY - JOUR ID - citeulike:5360516 L3 - citeulike-article-id:5360516 N2 - Computer aided diagnosis (CADx) systems for digitized mammograms solve the problem of classification between benign and malignant tissues while studies have shown that using only a subset of features generated from the mammograms can yield higher classification accuracy. To this end, we propose a mutual information-based Support Vector Machine Recursive Feature Elimination (SVM-RFE) as the classification method with feature selection in this paper. We have conducted extensive experiments on publicly available mammographic data and the obtained results indicate that the proposed method outperforms other SVM and SVM-RFE-based methods. IS - 16 JF - Pattern Recognition Letters SN - 01678655 EP - 1495 TI - Mutual information-based SVM-RFE for diagnostic classification of digitized mammograms VL - 30 SP - 1489 KW - feature_selection AU - Yoon, Sejong AU - Kim, Saejoon PY - 2009/12/01/ UR - http://dx.doi.org/10.1016/j.patrec.2009.06.012 ER -