Breast cancer diagnosis based on feature extraction using a hybrid of K-means and support vector machine algorithms
作者：
Highlights：
• Malignant and benign tumor abstract patterns are explored by K-SVM respectively.
• Similarities of tumors and abstract patterns is used for prediction model training.
• K-SVM reduces feature space dimensions significantly.
• Based on the WDBC dataset, the prediction model accuracy was at 97.38% by K-SVM.
• K-SVM saves the training time dramatically without losing prediction accuracy.
摘要
•Malignant and benign tumor abstract patterns are explored by K-SVM respectively.•Similarities of tumors and abstract patterns is used for prediction model training.•K-SVM reduces feature space dimensions significantly.•Based on the WDBC dataset, the prediction model accuracy was at 97.38% by K-SVM.•K-SVM saves the training time dramatically without losing prediction accuracy.
论文关键词：Data mining,K-means,Support vector machine,Cancer diagnosis
论文评审过程：Available online 30 August 2013.
论文官网地址：https://doi.org/10.1016/j.eswa.2013.08.044
