摘要
大多数模糊k-modes的相关改进算法仅关注对象之间的距离,并未关注对象的空间分布对于聚类的影响。将距离和密度双度量的测度方法引入模糊k-modes算法进行改进,该方法将对象的空间分布考虑在内,从而以一种更加合理的方式更新对象的隶属度。通过来自于UCI机器学习库的数据集测试算法改进前与改进后的性能,算法改进后的聚类正确率高于改进前的,证明算法改进后性能更好。
This paper introduces the double measure means of distance and density into fuzzy k-modes algorithm to im-prove it＇s performance.The method takes account into spatial distribution of data objects,with a more proper way to update the degree of membership of data objects.Using the datasets come from UCl machine learning repository to test the per-formance of the former and improved algorithm.The correct rate of improved algorithm is higher than the former,which shows the performance of improved algorithm is better than the former.
出处
《工业控制计算机》
2015年第9期90-91,94,共3页
Industrial Control Computer
基金
陕西省自然科学基金项目(2014JM2-6114)
陕西省教育厅自然科学专项(12JK0743)
教育部教改项目(2-3-ZXM-09)
