论文部分内容阅读
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 user experience of the social network, a personalized location recommendation service model is proposed, which considers the user’s sign-in behavior, the user’s characteristics and the semantic features of the location’s point of interest, and combines the ant colony algorithm with the improved hybrid The filtering algorithm can effectively combine the personalized location recommendation to improve the quality and efficiency of the personalized location recommendation.The experimental results show that the proposed location recommendation model has significantly better recall, accuracy and average absolute error than the existing methods .