NMF-based Method of Text Classification

被引：1

作者：

Sun, Fuzhen ^{[1
]}

Zhang, Kun ^{[2
]}

机构：

[1] Shandong Univ Technol, Coll Comp Sci & Tehnol, Zibo, Shandong, Peoples R China

[2] Nanyang Normal Univ, Dept Comp & Informat Technol, Nanyang, Henan, Peoples R China

来源：

2010 8TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION (WCICA) | 2010年

关键词：

NMF; text classification; conceptual semantic space; SVD;

D O I：

10.1109/WCICA.2010.5554018

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper put. forward a text classification method based on NMF (Non-negative Matrix Factorization), NMF-based analysis of the conceptual semantic space and the text vector dimensionality reduction, better explain the concept of the semantic vector, better reflect the local features of the text, comparison of two methods of generating conceptual semantic space based on NMF and SVD (Singular Value Decomposition). Experimental results show that the local conceptual semantic space generated based on NMF can be a better text classification accuracy.

引用

页码：4312 / 4316

页数：5

共 5 条

[1] DEERWESTER S, 1990, J AM SOC INFORM SCI, V41, P391, DOI 10.1002/(SICI)1097-4571(199009)41:6<391::AID-ASI1>3.0.CO
[2] 2-9
[3] Toward Faster Nonnegative Matrix Factorization: A New Algorithm and Comparisons
Kim, Jingu
Park, Haesun
[J]. ICDM 2008: EIGHTH IEEE INTERNATIONAL CONFERENCE ON DATA MINING, PROCEEDINGS, 2008, : 353 - 362
[4] Robust linear dimensionality reduction
Koren, Y
Carmel, L
[J]. IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, 2004, 10 (04) : 459 - 470
[5] PAUL S, 2003, P SPIE 03, P321

← 1 →