Research On Classification Of Decision Tree

The arrival of the Information Age, make people's lives more rich and colorful, also bringing a flood of data. Currently the problem that people hold large amounts of data but do not get any useful knowledge, in the situation of "data rich but information poor". How to effectively use the vast amount of data is the Information age gives us the most important issue. Data mining as the most effective means of a solution to the current explosion of data, lack of knowledge. In recent years has been rapid development.Decision tree of data mining technology is a very pop area of the mining method. Have many advantages different from other mining methods, many companies at home and abroad use decision tree as the preferred development of the mining means mining system. This paper analysis the existing decision tree classification algorithms and finds that these algorithms based on variable precision rough set have better classification accuracies and can tolerate the noise data. But when constructing decision tree based on variable precision rough set, these algorithms have the following shortcomings:the choice of attribute is difficult and the decision tree classification accuracy is not high. Therefore, this paper proposes a new variable precision rough set based decision tree algorithm, that is, weighted number of limits specific area (LVPER), the algorithm uses a new standard of attribute selection which considers comprehensively the classification accuracy and number of attribute values, that is, weight accuracy. At the same time support and confidence are introduced in the conditions of the corresponding node to stop splitting, and they can improve the algorithm's generalization ability. To reduce the impact of noise data and missing values, LVPER use the label predicted method based of match. Finally generate a complete model of decision of decision construction. And use the UCI experimental data test the superiority.This paper combined with the provincial department of education project" Renal failure, based on data mining of clinical hemodialysis treatment research the law", detailed analysis the characteristics of renal failure clinical data, proposed medical data preprocessing method. And affiliate the Hospital of Medical University and the clinical data available pretreatment and analysis of mining, a set of complete project development process. Trial of the experimental data mining, medical research combined with practical knowledge of the results of scientific proof, but also demonstrates the feasibility of the project. For the future a more extensive and in-depth medical data mining provides a complete set of mining ideas.This article is intended to improve the theory and practice of the combination, highlighting the theory of practical value in practice.