位置社交网络的个性化位置推荐

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.

Individual Location Recommendation for Location-Based Social Network

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 com-bines 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, ac-curacy and average absolute error value of the location recommendation model proposed in this article is superior to the existing methods.