The data mining method based on second learning

Decision tree algorithm is not only the important part of machine learning, but also the most widely used data mining tool. At present, there are many algorithms of generating decision tree, but when the database which we rely on exists noise, high quality knowledge is hard to obtain by ID3 algorithm. In this paper, we propose the data mining method based on second learning in case of ID3 algorithm, and analyze the performance of our method by a concrete database. Theory analysis and simulation indicate that this method posses the feature of strong operability, and it can improve the reliability of obtained knowledge.