Semi-supervised Cluster Ensemble based on Normal Mutual Information

摘要：

Semi-supervised cluster ensemble combines semi supervised learning and cluster ensemble to increase the preci sion of clustering. And it conquers the disadvantages of the selectivity, bias, low precision and bad generalization. And conquers that unsupervised learning cant take advantages of known information of datasets. As a result the precision, robustness and stability of cluster ensemble are increasing. Now semi-supervised cluster ensemble is researched at the very outset. The papers main goal is to construct an algorithm of semi-supervised cluster ensemble. In this paper it includes as follows. First the advantages of semi-supervised cluster ensemble are stated; Second the algorithm of semi-supervised cluster ensemble is established ; Third the experiments of semi supervised cluster ensemble, such as the results.the datasets, the types and the steps of experiments, are studied.

关键词： Semi-supervised Learning Cluster Ensemble Semi-supervised Cluster Ensemble

作者: Hongjun Wang Dezhi Yang Jianhuai Qi

作者单位: Information Research Institute South West Jiaotong University Chengdu, China PetroChina Planning and Engineering Institute PetroChina Company Limited,Beijing China

会议类型: 国际会议

会议名称: 2010 International Conference on Information Security and Artificial Intelligence(2010年信息安全与人工智能国际会议 ISAI 2010)

会议地点: 成都

会议语种:英文

页码: 1892-1894

在线出版日期: 2010-12-17（万方平台首次上网日期，不代表论文的发表时间）