摘要
在模糊ID3算法中,用模糊分类熵选择扩展属性,以自顶向下的方式递归地构建模糊决策树,对数据进行分类。提出了一种基于属性模糊熵的模糊分类算法,不同于模糊ID3算法,模糊条件属性的模糊熵作为权值用来对相对模糊频率进行加权,综合考虑各个模糊条件属性对分类的贡献。实例分析和实验结果表明了这一算法的有效性。
Fuzzy ID3 uses the fuzzy classification entropy as the criterion to select the expanded attributes,the fuzzy decision tree used for classification of data is recursively generated from top to bottom.Based on fuzzy entropy,this paper presents a fuzzy classification algorithm.Different from fuzzy ID3,the fuzzy entropies of fuzzy condition attributes are used as weights to weigh the relative fuzzy frequencies.The proposed algorithm integrates the contributions of all fuzzy condition attributes together.An illustrative example as well as the experimental results demonstrates the effectiveness of the proposed method.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第20期176-180,共5页
Computer Engineering and Applications
基金
国家自然科学基金No.60903088
河北省自然科学基金No.F2008000323,No.F2008000635,No.F2009000227
河北省应用基础研究重点项目(No.08963522D)
河北省教育厅科学研究计划项目No.2009312,No.2009410~~
