Popularity-aware collective keyword queries in road networks

Abstract

This paper addresses a popularity-aware collective keyword (PAC-K) query in road networks. Given a road network with POIs (Points of Interest), which is modeled as a road network graph, where each node locating in a two-dimensional space represents a road intersection or a POI, and each edge with weight represents a road segment, the PACK query aims to find a group of popular POIs (i.e., a popular region) that cover the query’s keywords and satisfy the distance requirements from each node to the query node and between each pair of nodes, such that the sum of rating scores over these nodes for the query keywords is maximized. We show the problem of answering the PACK query is NP-Hard. To solve this problem, we present exact and heuristic solutions on small and large road networks, respectively. In particular, to improve query performance, we propose a rating score scaling technique to reduce the search space and a redundant computation reducing technique to reduce the excessive redundant computations in query processing. Extensive performance studies using two real datasets confirm the efficiency, accuracy, and scalability of the proposed solutions.