摘要
模糊决策树推理是机器学习领域中的一种解决分类问题的有效算法,模糊推理方法的选择在很大程度上影响推理的性能和效果.对Min-Ambiguity,Fuzzy ID3和加权模糊决策树3种推理机制进行了对比研究,选择了推理过程中4种常用的算子(∨,∧)、(∨,×)、(+,∧)和(+,×)进行了对比分析,并在理论分析和实验验证的基础上提出了优先选择乘法算子(+,×)和(∨,×)的建议.
Fuzzy decision tree induction is an efficient algorithm for resolving classification problems in machine learning field. The choice of inductive method plays an important part in performance of fuzzy reasoning. Three heuristic algorithms, Min-Ambiguity, Fuzzy ID3 and the weighted fuzzy induction are comparatively introduced. Four matching operators, （∨,∧）、（∨,×）、（＋,∧）and（＋,×）, which are frequently used for applying fuzzy rules for classification are analyzed and compared. In the end, suggestion on how to choose the matching operators from experiments is developed.
出处
《河北大学学报（自然科学版）》
CAS
北大核心
2008年第4期433-437,共5页
Journal of Hebei University(Natural Science Edition)
关键词
模糊决策树
匹配算子
测试精度
Fuzzy decision tree
matching operator
testing accuracy
