Extension of similarity measures in VSM: From orthogonal coordinate system to affine coordinate system

摘要: Similarity measures are the foundations of many research areas, e.g. information retrieval, recommender system and machine learning algorithms. Promoted by these application scenarios, a number similarity have been proposed proposing. In state-of-the-art measures, vector-based representation is widely accepted based on Vector Space Model (VSM) in which an object represented as vector composed its features. Then, between two objects evaluated operations corresponding vectors, like cosine, extended jaccard, dice so on. However, there assumption that features independent each others. This apparently unrealistic, normally, relations features, i.e. co-occurrence keywords text mining area. this paper, space geometry-based method to extend VSM from orthogonal coordinate (OVSM) affine (AVSM) OVSM proved be special case AVSM. Unit vectors AVSM inferred considered angles unit vectors. At last, five different using Within numerous fields task clustering selected evaluation criterion. Documents AVSM, respectively. The results show outweighs OVSM.