期刊名称：International Journal of Hybrid Information Technology
印刷版ISSN：1738-9968
出版年度：2015
卷号：8
期号：4
页码：193-200
DOI：10.14257/ijhit.2015.8.4.22
出版社：SERSC
摘要：In this study, we used three popular data mining techniques (decision trees, artificial neural networks and support vector machine) to analyze the risk factors of medical expense of patients with hepatitis A in Guangdong Province in 2008. We compared the three methods to find out an effective method to predict medical expense and extract main influence factors of the medical expense. The results showed that support vector machine is the most accurate predictor.
关键词：medical expense; data mining; support vector machine
