摘要
将流形距离作为样本间相似性的基本度量测度,加入成对约束信息,通过近邻传播得出新的度量矩阵。把聚类问题转化为一优化数学模型。采用克隆选择算法求解这个优化模型,得出最后的聚类结果,通过人工数据集和UCI标准数据集验证了这种方法具有较高的准确性。
Manifold distance was used as the basic measure of the sample similarity between samples.The pair-wise constrains prior information was introduced,then the measure matrix was obtained through affinity propagation.So the clustering problem was transformed as one optimal model.Clonal selection algorithm was employed to solve this model,and the clustering results were given.Experiments on artificial data sets and UCI benchmark data set show that the proposed method can give the better accuracy.
出处
《计算机科学》
CSCD
北大核心
2012年第11期204-207,共4页
Computer Science
基金
吉林省自然科学基金项目(201215165)
符号计算与知识工程教育部重点实验室开放基金项目(93K-17-2010-K05)资助
关键词
流形距离
半监督聚类
人工免疫算法
Manifold distance
Semi-supervised clustering
Artificial immune algorithm
