基于决策树的数据挖掘算法优化研究

决策树模型是数据挖掘中最常用的一种方法，具有较好的分类预测能力，并能方便提取决策规则。基于相似性原理，以测试属性和决策属性的相似度作为启发规则构建决策树。提出了一种新的决策树生成算法。并在高校教师综合考评系统中采用了这种新算法，实验结果表明这种新的决策树生成算法预测精度较高，计算也比较简便。

Research on Optimizing Data Mining Algorithms Based on Decision Tree

The decision tree model is the most frequently used in the data mining, it is good classification and prediction＇s ability, and it can be convenient to draw the decision rule. Takes the similar degree of attribute between the test and the decision as the inspiring rule to produce the deci- sion tree,and then presents a new algorithm. In the system of the university teacher＇s synthetic comparison, we use the new algorithm to build a decision tree. And the experimental result in- dicates that the forecast precision of the new algorithm is better, and the computation is sim- pler.