LOCALLY REGULAR EMBEDDING

摘要：

Introducing the topological structure and regular topology structure, the purpose is to seek with regular topological structure of low dimensional data set, the structural topological structure regularity, and puts forward the measure to keep data set topology structure of local rules embedding method. Compared to nuclear feature mapping methods, such as Locally Linear Embedding, Laplacian Eigenmap and so on, low dimensional embedded result is approximately regular, and data classification has more natural connection. The last results prove the theory results show that this technique can greatly discover the topological structure of data, compared to the LLE and Laplacian Eigenmap.

关键词： topological structure regular topological structure Laplacian Eigenmap

作者: LuTan Yanrong Chi

作者单位: The institute of statistics & mathematics Shandong University of Finance Jinan, China

会议类型: 国际会议

会议名称: 2011 Seventh International Conference on Natural Computation(第七届自然计算国际会议 ICNC 2011)

会议地点: 上海

会议语种:英文

页码: 2171-2174

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）