頁籤選單縮合
題名 | A New Method for Constructing Decision Tree Based on GM(1,N) |
---|---|
作者姓名(外文) | Jane, Chuen-jiuan; Lee, Hung-chih; Lin, Yu-cheng; Chang, Ting-cheng; | 書刊名 | Journal of Grey System |
卷期 | 11:1 2008.03[民97.03] |
頁次 | 頁49-58 |
分類號 | 494.542 |
關鍵詞 | Decision tree; Grey theory; GM(1,N); Classification; |
語文 | 英文(English) |
英文摘要 | Decision Trees induction algorithms have been used for classification in a wide range of application domains. In the process of constructing a tree, the criteria of selecting test attributes will influence the classification accuracy of the tree. In this paper, our study aimed to construct a decision tree by adopting GM (1,N) model of Grey Theory to help decide which attributes should be eliminated to prevent the decision tree from overfitting. Finally, this study conducted an experiment whose results indicated that the proposed model outperforms the traditional decision tree. The result of an example shows that compared with the entropy-based approach, our approach is a better way to select nodes for constructing decision trees. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。