当前位置: X-MOL 学术Distrib. Parallel. Databases › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Social-aware spatial keyword top-k group query
Distributed and Parallel Databases ( IF 1.5 ) Pub Date : 2020-05-08 , DOI: 10.1007/s10619-020-07292-0
Xiangguo Zhao , Zhen Zhang , Hong Huang , Xin Bi

With the increasing popularity of location-based social networking services, information in social networks has become an important basis for analyzing user preferences. However, the existing spatial keyword group query only focuses on the distance constraint between the user groups, and ignores the social relationship between the user and his friends, which may affect the query results. Therefore, in order to meet the diverse query needs of user groups and improve user satisfaction based on information in social networks, this paper proposes a social-aware spatial keyword top-k group query problem. This problem aims to retrieve a set of k groups of POI objects that satisfy the preferences of multiple users, taking into account spatial proximity, social relevance, and keyword constraints. To solve this problem, we first design a rank function to measure the correlation between the query set and the candidate set. Next, in order to improve the query efficiency, we develop a novel hybrid index structure, SAIR-tree, which comprehensively considers the attributes of social, spatial, and textual. Then, we propose an approximate algorithm and an exact algorithm, combining with the pruning strategy, can efficiently search the top-k result set. Finally, experiments on real dataset confirm the efficiency and accuracy of the proposed algorithms.

中文翻译:

社会感知空间关键词top-k组查询

随着基于位置的社交网络服务的日益普及,社交网络中的信息已成为分析用户偏好的重要依据。然而,现有的空间关键词组查询只关注用户组之间的距离约束,而忽略了用户与好友之间的社交关系,可能会影响查询结果。因此,为了满足用户群体多样化的查询需求,提高基于社交网络信息的用户满意度,本文提出了社交感知空间关键词top-k群体查询问题。该问题旨在检索满足多个用户偏好的一组 k 组 POI 对象,同时考虑空间邻近性、社会相关性和关键字约束。为了解决这个问题,我们首先设计一个秩函数来衡量查询集和候选集之间的相关性。接下来,为了提高查询效率,我们开发了一种新的混合索引结构 SAIR-tree,它综合考虑了社会、空间和文本的属性。然后,我们提出了一种近似算法和一种精确算法,结合剪枝策略,可以有效地搜索top-k结果集。最后,在真实数据集上的实验证实了所提出算法的效率和准确性。结合剪枝策略,可以高效搜索top-k结果集。最后,在真实数据集上的实验证实了所提出算法的效率和准确性。结合剪枝策略,可以高效搜索top-k结果集。最后,在真实数据集上的实验证实了所提出算法的效率和准确性。
更新日期:2020-05-08
down
wechat
bug