| 摘要: |
| 首先介绍了ID3算法的基本思想，然后讨论了决策树算法中的难点问题，针对ID3算法中所存在的不足，提出了一种利用优值法的思想来改进信息增益的算法，并且与ID3算法进行了实验对比。实验表明，这种方法从树的规模和分类精度都优于许多决策树算法，使决策效率明显提高。 |
| 关键词: ID3算法 决策树算法 改进算法 信息增益 分类精度 决策效率 规模 基本思想 难点问题 |
| DOI：10.3969/j.issn.1001-893X. |
|
| 基金项目: |
|
| An Improved Algorithm Based on Decision Tree |
|
| () |
| Abstract: |
| This paper first introduces the main thought of ID3 algorithm, then discusses some difficult problems of selecting value on division in Decision Tree,and puts forward an algorithm to improve the gain of information, and makes a comparison with ID3 by experiment. Results demonstrate this algorithm is better than the others in the size of decision tree and classification accuracy. |
| Key words: Data Mining,Decision tree,Algorithm |
《电讯技术》编辑部 