Improving Of Decision Tree Algorithm Based On Discrete Degree

Classification mining is one of the most important techniques in data mining, as well as an important topic in the study. The point of the research orientation is the decision tree method. At present, a large number of scholars have advanced a lot of algorithm using decision tree to assort for cosmical data aggregation, but these algorithm are all improvements from different angles based on ID3 algorithm, which still has a lot of problems. Therefore, the paper combines decision tree algorithm with fuzzy set theory, and proposes a fuzzy aposteriori study algorithm. In this way, it can reduce error degree to assort due to indurations of fields value. Furthermore, we have advanced a sort of improving decision tree systematic algorithm-decision tree algorithm based on discrete degree. The algorithm not only improve right ratio to assort but it is also very effective.Lastly, we have used experiments to validate that improving decision tree algorithm is more effective than classical ID3 algorithm at capability.