Intrinsic dimension estimation based on maximum likelihood estimator and reverse k nearest neighbors

摘要：

We propose an algorithm for estimating the intrinsic dimension of a data set derived by applying the maximum likelihood estimator and analyzing both k nearest neighbor and reverse k nearest neighbors .Which can overcome the limitations of shortcut problem and bias caused by abnormal sampling density distribution when using only k nearest neighbor .It produces good results on some simulated and real datasets.

关键词： dimension estimation manifold learning visualization

作者: Guoming Chen Yiqun Chen Weiheng Zhu Jian Yin Nian Zhang

作者单位: Department of Computer Science Guangdong University of Education Guangzhou, GuangDong 510310. China Department of Computer Science Guangdong University of Education Guangzhou, GuangDong 510310, China Department of Computer Science JiNan University Guangzhou, GuangDong 510632, China Department of Computer Science Sun Yat-sen University Guangzhou, GuangDong 510275, China

会议类型: 国际会议

会议名称: 2011 Eighth International Conference on Fuzzy System and Knowledge Discovery(第八届模糊系统与知识发现国际会议 FSKD 2011)

会议地点: 上海

会议语种:英文

页码: 1342-1346

在线出版日期: 2011-07-26（万方平台首次上网日期，不代表论文的发表时间）