摘要
随着互联网与数据处理技术的迅速发展,网络信息的与日俱增,增加了用户从网络中快速获取有用信息的难度,而个性化推荐可以根据用户的自身属性与历史行为数据,为其推荐可能感兴趣的信息或商品,对人们日常生活产生了深远影响。本文在基于证据理论的置信协同推荐算法的基础上,改进组合规则,建立改进置信推荐模型,选取Epinions评分数据进行测试,并对不同改进模型的准确度进行检验。
With the rapid development of the Internet and data processing technology,the amount of network information increases.Con⁃sequently users are more difficult to obtain useful information quickly from the network.Personalized recommendations can suggest in⁃formation or goods for users Based on their own attributes and historical behavior data,which have a profound impacton our daily life.In this paper,a new evidentialcollaborativerecommendation algorithm is developed in the framework of belief functions.The combination rules are improved to in the model.The Epinions score data are used for test and the accuracy of different improved models are compared.
作者
马丽娜
MA Li-na(School of Economic and Statistics,Xingzhi College of Xi'an University of Finance and Economics,Xi’an 710038,China)
出处
《电脑知识与技术》
2019年第12Z期207-208,211,共3页
Computer Knowledge and Technology
基金
陕西省教育厅专项科学研究计划(19JK0330)
关键词
推荐系统
证据理论
不确定评分
Recommendation system
Evidence theory
Soft rating
