| 基于粗糙集决策树算法的研究 |
| |
| 引用本文： | 彭莉芬,陈俊生,胡学钢.基于粗糙集决策树算法的研究[J].安庆师范学院学报(自然科学版),2012,18(1):75-78. |
| |
| 作者姓名： | 彭莉芬 陈俊生 胡学钢 |
| |
| 作者单位： | 安徽电子信息职业技术学院,安徽蚌埠233000／合肥工业大学,安徽合肥230009;合肥工业大学,安徽合肥,230009 |
| |
| 摘 要： | 传统的决策树方法在实际应用中存在很多不足,如生成树规模过大,抗噪性较差等,因此,提出了将变精度粗糙集和混合变量集算法应用于决策树分类中,通过变精度和混合属性集分类减小树的规模和过度拟合问题,降低了噪声数据对属性选择的影响,并通过实验证明该算法与传统的算法相比具有较大的优越性。
|
| 关 键 词： | 决策树 粗糙集 变精度 混合变量集 |
Research on Decision Tree Algorithm Based on Rough Set |
| |
| PENG Li-fen,CHENG Jun-sheng,HU Xue-gang.Research on Decision Tree Algorithm Based on Rough Set[J].Journal of Anqing Teachers College(Natural Science Edition),2012,18(1):75-78. |
| |
| Authors: | PENG Li-fen CHENG Jun-sheng HU Xue-gang |
| |
| Institution: | 1.Anhui Vocational college of Electronics and Information Technology,Bengbu,Anhui 233000,China; 2.Hefei University of Technology,Hefei,Anhui 230009,China) |
| |
| Abstract: | In practice,there are many deficiencies of the traditional decision tree method,such as spanning tree is too large,poor noise resistance.In this paper,the variable precision rough set and mixed set of variables are used in decision tree classification algorithm,by varying the mixed attribute set classification accuracy and variable precision rough set reduced the size of the tree and over-fitting problem,reduced the noise data on the impact of feature selection,The experiment shows that the algorithm has great advantages compared with the traditional method. |
| |
| Keywords: | decision tree rough set variable precision mixed attribute set |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
