Chao Cheng-Min, Yu Ya-Wen, Cheng Bor-Wen, Kuo Yao-Lung
Department of Business Administration, National Taichung University of Science and Technology, Taichung, Taiwan.
J Med Syst. 2014 Oct;38(10):106. doi: 10.1007/s10916-014-0106-1. Epub 2014 Aug 14.
The aim of the paper is to use data mining technology to establish a classification of breast cancer survival patterns, and offers a treatment decision-making reference for the survival ability of women diagnosed with breast cancer in Taiwan. We studied patients with breast cancer in a specific hospital in Central Taiwan to obtain 1,340 data sets. We employed a support vector machine, logistic regression, and a C5.0 decision tree to construct a classification model of breast cancer patients' survival rates, and used a 10-fold cross-validation approach to identify the model. The results show that the establishment of classification tools for the classification of the models yielded an average accuracy rate of more than 90% for both; the SVM provided the best method for constructing the three categories of the classification system for the survival mode. The results of the experiment show that the three methods used to create the classification system, established a high accuracy rate, predicted a more accurate survival ability of women diagnosed with breast cancer, and could be used as a reference when creating a medical decision-making frame.
本文旨在运用数据挖掘技术建立乳腺癌生存模式分类，并为台湾地区被诊断为乳腺癌的女性的生存能力提供治疗决策参考。我们研究了台湾中部一家特定医院的乳腺癌患者，以获取1340个数据集。我们采用支持向量机、逻辑回归和C5.0决策树来构建乳腺癌患者生存率的分类模型，并使用10折交叉验证方法来识别该模型。结果表明，为模型分类建立的分类工具的平均准确率均超过90%；支持向量机为构建生存模式分类系统的三类提供了最佳方法。实验结果表明，用于创建分类系统的三种方法具有较高的准确率，能够更准确地预测被诊断为乳腺癌的女性的生存能力，并且在创建医疗决策框架时可作为参考。
