A Classification Prediction Method using Rough Set and Decision Tree
Shangzhi Wu, Xia Ning, Yixuan Ren, Zhining Wang · 2022
In the process of data classification prediction, the data collected in reality are high dimensions and there are redundant attributes leading to large deviation of results. A classification prediction method based on rough set and decision tree is proposed, which integrates genetic algorithm and ten-fold cross validation method. In this method, the core attribute is calculated by the attribute dependence of rough set, and the condition attribute with the strongest decision attribute dependence is found by the global optimization of the core attribute through genetic algorithm. Then, the decision tree is constructed for the subset of the condition attributes, and the ten-fold cross validation is used to avoid the over-fitting phenomenon in the process of decision tree construction. The experimental results show that the decision tree classifier constructed by this method compared with other classifiers, the accuracy of classification prediction is improved effectively.