Efficient Processing of Group Planning Queries Over Spatial-Social Networks

Recently, location-based social networks, that involve both social and spatial information, have received much attention in many real-world applications such as location-based services (LBS), map utilities, business planning, and so on. In this paper, we seamlessly integrate both social networks and spatial road networks, resulting in a so-called spatial-social network, and study an important and novel query type, named group planning query over spatial-social networks (GP-SSN), which is very useful for applications such as trip recommendations. In particular, a GP-SSN query retrieves a group of friends with common interests on social networks and a number of spatially close points of interest (POIs) on spatial road networks that best match group's preferences and have the smallest traveling distances to the group. In order to tackle the GP-SSN problem, we design effective pruning methods, matching score pruning, user pruning, and distance pruning, to rule out false alarms of GP-SSN query answers and reduce the problem search space. We also propose effective indexing mechanisms to facilitate the GP-SSN query processing, and develop efficient GP-SSN query answering algorithms via index traversals. Extensive experiments have been conducted to evaluate the efficiency and effectiveness of our proposed GP-SSN query processing approaches.

Automated summary

In this paper, the authors focus on the integration of social networks and spatial road networks to form a spatial-social network. They propose an important and novel query type called GP-SSN, which retrieves a group of friends with common interests on social networks and a number of spatially close points of interest on spatial road networks that best match the group's preferences. The authors design effective pruning methods and indexing mechanisms to solve the GP-SSN problem and develop efficient query answering algorithms.