Given a road network and a social network,the collective spatial keyword query aims to find a set of points of interest(POIs)in which the text information contains all query keywords close to the query location and with a small mutual distance.The query goal of the cohesive group is to identify a group of users that are closely connected geo-graphically and socially,whereas the query purpose of the collective spatial keyword cohesive group is to determine a pair of optimally matched POI sets and user sets that satisfy the query requirements.To address this problem,a novel type of cohesive group query mode is proposed for collective spatial keywords.Initially,the candidate POI set is ob-tained through a fast greedy query process.Then,the core tree structure is used to store the results of(k,c)-core decom-position to improve the efficiency of cohesive group query and ensure that the query results can satisfy the social con-straints among users and the spatial constraints among POIs simultaneously.The experiments conducted on real data-sets show that the proposed method is one to two orders of magnitude faster than the query efficiency of the enumera-tion method,and the results exhibit high query accuracy.
维普
万方数据