摘要
非负矩阵分解是一种约束矩阵元素非负的矩阵分解技术。非负矩阵分解将高维的数据矩阵分解成为低维的基矩阵和系数矩阵,解决数据压缩与聚类等数据挖掘任务。非负矩阵分解在机器学习、图像处理等领域得到广泛应用,未来仍有较大的发展空间。本文对非负矩阵分解方法的发展历史、算法和实际应用等方面进行阐述,并分析非负矩阵分解现有的不足,指出进一步研究的方向。
Non-negative matrix factorization is a matrix factorization technique that constrains matrix elements to be nonnegative.Non-negative matrix factorization decomposes high-dimensional data matrix into low-dimensional base matrix and coefficient matrix to solve data mining tasks such as data compression and clustering.Non-negative matrix factorization is widely used in machine learning,image processing and other fields,and there is still a lot of room for development in the future.This article expounds the development history,algorithm and practical application of non-negative matrix factorization,analyzes the existing shortcomings of non-negative matrix factorization,and points out the direction of further research.
作者
王宇辰
WANG Yu-chen(School of Statistics,Lanzhou University of Finance and Economics,Lanzhou Gansu 730020)
出处
《数字技术与应用》
2021年第2期112-114,共3页
Digital Technology & Application
关键词
非负矩阵分解
数据挖掘
聚类算法
Non-negative matrix factorization
Data mining
Clustering algorithm
