基于非负矩阵分解与调和函数的半监督学习方法  

Semi-supervised Learning Approach Based on Non-negative Matrix Factorization and Harmonic Functions

作　　者：张帆[1] 李林 Zhang Fan;Li Lin(College of Arts and Sciences,National University of Defense Technology,Changsha 410073,China;Centre for Assessment and Demonstration,Academy of Military Science,Beijing 100091,China)

机构地区：[1]国防科技大学文理学院,长沙410073 [2]军事科学院评估论证研究中心,北京100091

出　　处：《统计与决策》2019年第22期71-73,共3页Statistics & Decision

基　　金：国防科技大学优秀研究生创新资助项目(4345133214);军事科学院评估论证研究中心青年基金项目

摘　　要：数据分类是统计学的重要组成部分,为统计与决策提供高质量的数据支撑。半监督学习作为重要的数据分类方法,近年来被广泛研究,文章提出了一种基于非负矩阵分解与调和函数的半监督学习方法(NMF-HF)。首先,对原始数据矩阵进行非负矩阵分解(NMF)从而得到特征矩阵,利用调和函数(HF)的性质对原始数据进行数据分类,最后将该方法(NMF-HF)与经典分类方法进行分类准确率对比,得到的实验结果证明了NMF-HF的高效性。Data classification is an important part of statistics, which provides high-quality data support for statistics and decision-making. As an important data classification, semi-supervised learning has been widely studied in recent years. In this paper,a novel approach based on non-negative matrix factorization and harmonic functions(NMF-HF) is proposed for semi-supervised learning. Firstly, the feature data is extracted from the original data by non-negative matrix factorization(NMF), and then the original data is classified by harmonic functions(HF) on the basis of the feature data. Finally, the classification accuracy of this method(nmf-hf) is compared with that of the classical classification method. Empirical results demonstrate the high efficiency of NMF-HF.

关 键 词：非负矩阵分解 调和函数 分类准确率 半监督学习 

分 类 号：O21[理学—概率论与数理统计]