Text representation: from vector to tensor

Author

Liu, Ning ; Zhang, Benyu ; Yan, Jun ; Chen, Zheng ; Liu, Wenyin ; Bai, Fengshan ; Chien, Leefeng

Author_Institution

Dept. of Math. Sci., Tsinghua Univ., Beijing, China

fYear

2005

fDate

27-30 Nov. 2005

Abstract

In this paper, we propose a text representation model, Tensor Space Model (TSM), which models the text by multilinear algebraic high-order tensor instead of the traditional vector. Supported by techniques of multilinear algebra, TSM offers a potent mathematical framework for analyzing the multifactor structures. TSM is further supported by certain introduced particular operations and presented tools, such as the High-Order Singular Value Decomposition (HOSVD) for dimension reduction and other applications. Experimental results on the 20 Newsgroups dataset show that TSM is constantly better than VSM for text classification.

Keywords

singular value decomposition; tensors; text analysis; vectors; dimension reduction; high-order singular value decomposition; multifactor structures; multilinear algebraic high-order tensor; tensor space model; text representation; vector space model; Asia; Computer science; Data mining; Indexing; Information retrieval; Large scale integration; Matrix decomposition; Principal component analysis; Singular value decomposition; Tensile stress;

fLanguage

English

Publisher

ieee

Conference_Titel

Data Mining, Fifth IEEE International Conference on

ISSN

1550-4786

Print_ISBN

0-7695-2278-5

Type

conf

DOI

10.1109/ICDM.2005.144

Filename

1565767