基于用户轨迹和好友关系的兴趣点推荐

A point of interest recommendation model based on tracks and friend relationship of users

连续兴趣点(POI)推荐是基于地理位置社交网络(LBSN)的重要应用之一,已有研究提出采用兴趣点信息和时空信息进行推荐的方法,但没有充分地利用相关辅助信息,因此无法解决用户短轨迹签到导致的信息不足问题。针对这些问题,提出一种整合好友关系和自注意力的兴趣点推荐模型ATFR。该预测模型包含3个部分:首先,通过图嵌入的方法得到好友关系的向量表示并利用GRU得到用户兴趣偏好向量;其次,利用自注意力机制对用户签到序列的顺序影响和社交影响建模,有选择地关注签入序列中相关的历史签入记录;最后,根据兴趣点排序列表进行未来兴趣点推荐。在2个真实数据集上的实验结果表明ATFR模型有更好的表现,可以用来提高网站应用和个性化兴趣点推荐服务的质量。

Point of interest(POI)recommendation is one of the most important applications of location-based social networking(LBSN).Existing studies have proposed methods that utilize POI information and spatio-temporal information for recommendation,but the relevant auxiliary information has not been fully utilized,so it can not alleviate the problems of insufficient information and less travel data caused by short track check-in of users.To solve these problems,a friends relationship and self-attention recommendation model ATFR is proposed.The prediction model consists of three parts:Firstly,the representation vector of friend relationship is obtained by graph embedding method and input into the neural network.Secondly,the user interest preference vector is obtained by GRU.Then,the self-attention mechanism is used to model the sequential and social effects of the user check-in sequence and selectively focus on the historical check-in records associated with the check-in sequence.Finally,the future interest points are recommended according to the interest points sorting list.Experimental results on two public datasets show that this model performs better than other algorithms and can be used to improve the quality of personalized interest recommendation services for websites and applications.