Mining a Small Medical Data Set by Integrating the Decision Tree and t-test

Abstract

Although several researchers have used statistical methods to prove that aspiration followed by the injection of 95% ethanol left in situ (retention) is an effective treatment for ovarian endometriomas, very few discuss the different conditions that could generate different recovery rates for the patients. Therefore, this study adopts the statistical method and decision tree techniques together to analyze the postoperative status of ovarian endometriosis patients under different conditions. Since our collected data set is small, containing only 212 records, we use all of these data as the training data. Therefore, instead of using a resultant tree to generate rules directly, we use the value of each node as a cut point to generate all possible rules from the tree first. Then, using t -test, we verify the rules to discover some useful description rules after all possible rules from the tree have been generated. Experimental results show that our approach can find some new interesting knowledge about recurrent ovarian endometriomas under different conditions.