摘要
为了有效改善位置社交网络的用户体验,提出了一种个性化位置推荐服务模型.综合考虑了用户的签到行为特点、用户特征及位置兴趣点的语义特征,并将蚁群算法与改进的混合协同过滤算法有效结合起来进行个性化位置推荐,以此提高个性化位置推荐的质量和效率.实验结果表明,提出的位置推荐模型的召回率、准确率和平均绝对误差值都明显优于已有方法.
In order to effectively improve the users＇ experience for location social networks,a model of personalized location recommendation service was proposed. Considering the users＇ check- in behavior features,the users＇ characteristics and semantic features of interested location point,this model combines the ant colony algorithm with the improved hybrid collaborative filtering algorithm to improve the quality and efficiency of the individual location recommendation. Experiments show that,the recall,accuracy and average absolute error value of the location recommendation model proposed in this article is superior to the existing methods.
出处
《北京邮电大学学报》
EI
CAS
CSCD
北大核心
2015年第5期118-124,共7页
Journal of Beijing University of Posts and Telecommunications
基金
国家自然科学基金项目(61370139)
网络文化与数字传播北京市重点实验室项目(ICDD201506)
北京市属高等学校创新团队建设与教师职业发展计划项目(IDHT20130519)
关键词
位置社交网络
个性化位置推荐
位置服务
协同过滤算法
location-based social network
individual location recommendation
location-based service
collaborative filtering
