基于聚类思想的加权条件熵及属性约简

The Weighted Conditional Entropy and Attribute Reduction Based on Clustering

通过聚类分析的方法得到信息粒与对象权重的确定方法,同时将对象权重与熵理论知识相结合定义了一种加权条件熵.最后基于新定义的加权条件熵得到一种改进的属性重要度确定方法和相应的属性约简算法,并且用UCI中的几组数据集验证了该算法的可行性和合理性.

Information granules and a method to determine the weight of the object were obtained by using clustering analysis.At the same time,a weighted conditional entropy was defined by combining the object weights with the entropy theory.Finally,an improved method to determine the attribute significance and the corresponding attribute reduction algorithm were obtained based on the new definition of weighted conditional entropy.Experiments were performed on UCI data sets.And the results showed that the proposed algorithm was feasible and reasonable.