摘要
针对推荐系统中数据量越来越大,其对应的矩阵填充问题算法效率有待提升.基于随机算法策略以及高效数据访问要求,提出一种新的求解矩阵填充问题的算法,并借助Matlab软件实现该算法.数值试验结果表明,该算法在效率上可提升30%左右.
In view of the problem that with increasing large amount of datas in a recommendation system,the computing efficiency of its corresponding matrix completion algorithms need to be improved.Based on randomized algorithm strategy and efficient data access requirements,a new algorithm for solving matrix completion problem was proposed,and the Matlab software was employed to realize the algorithm.The numerical experiment result shows that the new algorithm can speed up the computing efficiency of original one by about 30%.
作者
陈熙
邓杰臣
席势鸿
刘晓辰
张相君
冯月华
CHEN Xi;DENG Jiechen;XI Shihong;LIU Xiaochen;ZHANG Xiangjun;FENG Yuehua(School of Mechanical and Automotive Engineering,Shanghai 201620,China;School of Air Transportation and School of Flying,Shanghai 201620,China;School of Mathematics,Physics and Statistics,Shanghai University of Engineering Science,Shanghai 201620,China)
出处
《上海工程技术大学学报》
CAS
2021年第3期281-284,共4页
Journal of Shanghai University of Engineering Science
基金
上海工程技术大学大学生创新训练计划资助项目(CX2021008)。
关键词
奇异值分解
随机算法
低秩逼近
推荐系统
singular value decomposition(SVD)
randomized algorithm
low-rank approximation
recom-mendation system
