基于时空序列模式匹配的兴趣点推荐方法

POI recommendation method based on pattern matching of spatio-temporal sequence

兴趣点(point of interest,POI)是地理空间信息的重要组成部分,在基于位置的信息服务中被广泛使用。针对用户对兴趣点访问所产生的时空序列,利用闭合序列模式挖掘方法分析频繁模式,在此基础上根据用户当前所处位置或最近访问序列,通过序列分析进行时空序列模式匹配,并按照序列的匹配程度给出兴趣点推荐列表。实验结果表明,闭合序列模式挖掘与时空序列模式匹配相结合的方法能够有效地应用在兴趣点推荐中,有利于引导用户的兴趣点访问行为,从而提升位置服务的质量。

Point of interest（POI） is a major component of geospatial information,it is widely used in location based services.In this paper,closed sequential pattern mining method is used to analyze the frequent pattern of spatio-temporal sequence,sequence analysis method is performed for pattern matching,and POI recommendation list is generated according to the matching degree.The experimental results show that the combination method of closed sequential pattern mining and pattern matching can be applied in POI recommendation effectively,it is useful to guide the POI access behavior of users and thus improve the quality of location-based services.