西北师范大学学报（自然科学版）2018，Vol.54Issue(3)：51-56,6.DOI:10.16783/j.cnki.nwnuz.2018.03.010
基于矩阵奇异值分解的文本分类算法研究
Study on text classification algorithm based on singular value decomposition
摘要
Abstract
The weakness of KNN text classification algorithm is larger computational overhead which leads to low efficiency in high-dimensional dataset.A method of dimension reduction based on singular value decomposition(SVD)is adopted to realize the dimension reduction of the feature vector and remain more classification information.At the same time,information gain(IG)is used to filter the items that have almost no contribution to the classification system,w hich overcomes the problem that the computational overhead of SVD is rising w hen the dimension of feature vector is increasing.Experiments show that this method can reduce the computing cost effectively and keep higher precision of text classification.关键词
文本分类/奇异值分解/信息增益Key words
text classification/singular value decomposition/information gain分类
信息技术与安全科学引用本文复制引用
景永霞,王治和,苟和平..基于矩阵奇异值分解的文本分类算法研究[J].西北师范大学学报（自然科学版）,2018,54(3):51-56,6.基金项目
海南省自然科学基金资助项目(617160) （617160）
海南省高等学校科学研究项目(Hnky2015-72) （Hnky2015-72）
海南省高等学校教育教学改革研究项目(Hnjg2017-68) （Hnjg2017-68）
