Summary of decision tree algorithm and its application in attribute reduction

In this paper, for the refinement of the database in data mining, by synthetically analyzing the characteristics of the current attribute reduction methods and decision tree algorithm, we put forward formalized description model of rule knowledge, and establish a kind of attribute reduction method (BD-RED) of decision tree by using similarity between rules families. Further, we discuss the construction of similarity measure between rules families, and give the specific implementation strategy of BD-RED, then analyze the performance through examples. The results indicate that, BD-RED, with the features of good structure and strong operability, is an effective way to achieve attribute reduction under different decision consciousness, so it can be suitable for the large scale attribute reduction.