摘要
O2O模式已成为移动互联网环境下餐饮业的发展方向,如何向用户精准地推荐所需餐食成为餐饮服务平台关注的热点问题.协同过滤是目前餐饮推荐系统中运用最广泛的技术,但随着信息量的增加,协同过滤技术面临着数据集稀疏性和冷启动等问题.文章针对餐食数据矩阵的稀疏性问题,设计一种基于SVD技术协同过滤的餐饮推荐系统,实验表明该技术能够提升协同过滤算法的评估精度与评估效果.
The O2 O model has become the development direction of the catering industry in the mobile Internet environment.How to recommend catering for users is a hot issue of the catering service platform.At present,collaborative filtering is the most widely used technology in catering recommendation system.However,as the amount of information increases,collaborative filtering technology also faces the problems of data set sparsity and cold start.In this paper,a catering recommendation system based on SVD collaborative filtering is designed to solve the sparsity problem of the meal data matrix.The experiments show that the proposed technology can improve the evaluation accuracy and effect of collaborative filtering algorithms.
出处
《浙江树人大学学报（自然科学版）》
2018年第2期1-5,共5页
Journal of Zhejiang Shuren University(Acta Scientiarum Naturalium)
基金
2017年度浙江省大学生科技创新项目(2017R421027)。
关键词
协同过滤
个性推荐
矩阵稀疏性
SVD
collaborative filtering
personality recommendation
matrix sparsity
SVD
