基于 ID3 决策树算法的改进研究

Research on the Improvement of Decision Tree Algorithm Based on ID3

决策树是典型的归纳学习和数据挖掘方法,通过对数据库中获取的数据项属性值进行划分归类,最终形成类似于流程图的树型结构形式。ID3算法是决策树中的核心算法,针对ID3算法倾向于取值较多的属性的缺点,通过引入泰勒公式与麦克劳林公式,对传统算法进行降维,减小算法的计算复杂度,提高算法运行效率,使决策树的生成时间缩短,算法的效率得到了较大的提高。

Decision tree algorithm is a typical way of induction and data mining,which can be explained as a process to acquire data properties from database to classify and form a tree structure akin to the flow chart. ID3 is the core algorithm of decision tree which aims at the flaws of redundant values. This method in- troduces Taylor’s formula and Maclaurin’s formula to realize the dimension- reduction algorithm,lessen count-ing complexity,increase efficiency and shorten the generation of decision tree.