基于情景上下文与信任关系的旅游景点推荐算法  

Attractions recommendation algorithm based on situational context and trust relationship

作　　者：沈记全[1] 王磊[1] 侯占伟[1] 薛霄[1] Shen Jiquan;Wang Lei;Hou Zhanwei;Xue Xiao(College of Computer Science ＆ Technology,Henan Polytechnic University,Jiaozuo Henan 454000,China)

机构地区：[1]河南理工大学计算机科学与技术学院,河南焦作454000

出　　处：《计算机应用研究》2018年第12期3640-3643,共4页Application Research of Computers

基　　金：国家自然科学基金青年基金资助项目(61300124);河南省基础与前沿资助项目(152300410212);河南省科技攻关资助项目(162102310426;172102310250);河南省教育厅自然科学基金资助项目(17A520034)

摘　　要：针对现有旅游景点推荐个性化不足的问题,提出了一种基于信任关系与情景上下文的旅游景点推荐算法。首先在传统的协同过滤算法上以用户信任度代替相似度来解决数据稀疏性;其次引入用户情景上下文信息,更全面地反映出用户的个性化需求;最后基于用户的信任度和上下文信息优化,建立一个推荐结果准确度更高的旅游景点推荐模型。模拟实验结果表明,综合考虑信任度和情景上下文信息的推荐策略表现最优。In consideration of the lack of personalization in the recommendation algorithms of tourist attractions,this paper proposed a recommendation algorithm of tourist attractions based on trust relationship and context. Firstly,it used user trust instead of the similarity to solve the problem of data sparse based on the traditional collaborative filtering algorithm. Secondly,it used user scenario context information to reflect the personalized needs of users. Finally,it established a recommended model of tourist attractions with higher accuracy of recommendation results based on user trust and contextual information optimization. The result of simulation experiment shows that the recommendation strategy which takes trust and contextual context information into consideration has the better performance than those existed strategies.

关 键 词：情景上下文 信任度 协同过滤 推荐 个性化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]