Processing Of Multiple-user Spatial Queries

As the prosperous of smart terminal and location-based serviced,geosocial networking are closely related to people's daily life,such as food sourcing,location-based recommendation,ad hoc networking,etc.In geosocial networking,various kinds of queries are submitted by query users,such as range queries,top-k queries,skyline queries,etc.However,these queries only focus on a single query user associated with a location;they ignore the problem of a group of users.Further,in query processing,users' individual preferences are latent.These queries can can not evaluate their preferences accurately,returning imprecise result for query users.This calls for techniques that support processing of multiple-user location-based queries.In this paper,we only concentrate on Multiple-user Location-based Keyword queries,which return a set of Point-of-Interests(POIs)that are 'close' to the locations of the users in a group and can provide them with potential options at the lowest expenseWe mainly focus on Multiple-user Location-based Keyword queries.The road network is another key iseue while processing MULK queries in geosocial networking.Thus,we further study MULK queries on road network,and propose corresponding effective solutions.Our contributions are as follows:(1)To answer MULK queries,we present its formalized definition of MULK,and further design a dynamic programming-based algorithm to find the optimal result set.Since its execution time is intolerable for query users,we present two approximation algorithms(MULK Appro1 and MULK Appro2)to improve MULK query processing efficiency.Finally,we conduct a set of experiments that demonstrate the effectiveness and efficiency of our solutions under various parameter settings.(2)To mine users' preference in MULK queries,based on user preference weight matrix,we propose an Interactive Multiple User Location-based Keyword query,which mines the individual preferences of multiple users by interacting with them.(3)To index POIs in road network,we provide a hierarchical index of road networks,and based on this index,we propose a solution for processing MULK queries on road network efficiently.In addition,we present an algorithm for updating the index.