Crowds' Classification Using Hierarchical Cluster, Rough Sets, Principal Component Analysis and Its Combination

Abstract

13 kind of nationalities crowds' data classification using hierarchical cluster (HC), rough sets (RS), principal component analysis (PCA) and its combination, the result shows: first, rough sets and principal component analysis can dimensionality reduction and de-noising; second, hierarchical cluster after rough sets (RSHC), principal component analysis after rough sets (PCARS), principal component analysis after principal component analysis (PCAPCA), hierarchical cluster after principal component analysis (HCPCA), rough sets after principal component analysis (PCARS) are similarly result. Then, according to different practical application select different methods or combinative methods, which can maximize their advantages and minimize their disadvantages.