Research On Technology Of Data Mining Algorithm Based On Decision Tree

The decision tree learning algorithm is very useful in data mining technique. The ID3 algorithm is studied in this thesis, which is the key algorithm of decision tree learning. The improved algorithm of ID3 is proposed by introducing user-interest, which resolves the problem of "the big data covers up the fraction" in the decision support process in certain degree.Firstly, based on the traditional IDS, the method of improving IDS is put forward by introducing the concept of user-interest, which adds user-interest to information entropy in the IDS algorithm. Therefore it makes the algorithm depend not only on train sets but also on field knowledge of studied attribute. In this way the improved algorithm increases the classification precision. And an example is trained by IDS, which proves that improved algorithm is efficient.Secondly, the thesis puts forward that conditional probability of attribute to positive example can be used to compare the information which attribute provides so as toconstruct decision tree and to get classify rules. And a demonstration shows that the algorithm simplifies the decision tree's building process efficiently.To evaluate the algorithm, a data-mining trial system is designed which uses different algorithms to extract the decision rules in Northwind database. By this trial system practical commercial rules can be obtained.