基于时空轨迹相似性的兴趣点推荐算法  

Interest Point Recommendation Algorithm Based on Spatiotemporal Trajectory Similarity

作　　者：陶健 王睿[1] 殷西祥[1] TAO Jian;WANG Rui;YIN Xixiang(Anhui Business College of Vocational Technology,Wuhu 241002,China)

机构地区：[1]安徽商贸职业技术学院信息与人工智能学院,安徽芜湖241002

出　　处：《通化师范学院学报》2024年第10期55-66,共12页Journal of Tonghua Normal University

基　　金：安徽省自然科学重点研究项目(2022AH052741);安徽省高校优秀人才支持计划重点项目(gxypZD2020056);安徽商贸职业技术学院校级重点科学研究项目(2021ZDG05)。

摘　　要：针对现有兴趣点推荐中忽略用户轨迹间存在的依赖关系,未充分考虑轨迹间的时空间隔问题,该文提出一种基于时空轨迹相似性的兴趣点推荐算法G-IPRTS.该方法首先对轨迹预处理,将其转化为有意义的停留点轨迹.其次在停留点轨迹相似性处理中利用Geohash编码,以及时间维过滤提高轨迹相似性的精度和速度.同时在兴趣点推荐中引用熵权法构建兴趣点全局属性融合得分机制,与用户局部动态轨迹偏好相结合,进行Top-K兴趣点推荐.最后在两个真实数据集上进行实验比较与分析,实验结果表明该方法能够更好地提高轨迹相似性的处理,提升兴趣点推荐的性能.As to the problems of neglecting the dependency relationship between user trajectories in existing interest point recommendations and not fully considering the spatiotemporal interval between trajectories,this article proposes an interest point recommendation algorithm G-IPRTS based on spatiotemporal trajectory similarity.This method first preprocesses the trajectory and converts it into meaningful dwell point trajectories.Secondly,Geohash encoding and time dimension filtering are used to improve the accuracy and speed of trajectory similarity in the processing of dwell point trajectory similarity.At the same time,the entropy weight method is used in interest point recommendation to construct a global attribute fusion scoring mechanism for interest points,which is combined with user local dynamic trajectory preferences to perform Top-K interest point recommendation.Finally,experimental comparison and analysis were conducted on two real datasets,and the experimental results showed that this method can better improve the processing of trajectory similarity and the performance of interest point recommendation.

关 键 词：兴趣点推荐 停留点轨迹 轨迹相似性 Geohash编码 熵权法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]