Socio-Spatial Group Queries for Impromptu Activity Planning

Abstract

The development and integration of social networking services and smartphones have made it easy for individuals to organize impromptu social activities anywhere and anytime. Main challenges arising in organizing impromptu activities are mostly due to the requirements of making timely invitations in accordance with the potential activity locations, corresponding to the locations of and the relationships among the candidate attendees. Various combinations of candidate attendees and activity locations create a large solution space. Thus, in this paper, we propose Multiple Rally-Point Social Spatial Group Query (MRGQ), to select an appropriate activity location for a group of nearby attendees with tight social relationships. We first consider a special case of MRGQ, namely the Socio-Spatial Group Query (SSGQ), to determine a set of socially acquainted attendees while minimizing the total spatial distance to a specific activity location. We prove that SSGQ is NP-hard and formulate an Integer Linear Programming optimization model for SSGQ. We then develop an efficient algorithm, called SSGS, which employs effective pruning techniques to reduce the running time to determine the optimal solution. Moreover, we propose a heuristic algorithm for SSGQ to efficiently produce good solutions. We next consider the more general MRGQ. Although MRGQ is NP-hard, the number of attendees in practice is usually small enough such that an optimal solution can be found efficiently. Therefore, we first propose an Integer Linear Programming optimization model for MRGQ. We then design an efficient algorithm, called MAGS, which employs effective search space exploration and pruning strategies to reduce the running time for finding the optimal solution. We also propose to further optimize efficiency by indexing the potential activity locations. A user study demonstrates the strength of using SSGS and MAGS over manual coordination in terms of both solution quality and efficiency. Experimental results on real datasets show that our algorithms can process SSGQ and MRGQ efficiently and significantly outperform other baseline algorithms, including one based on the commercial parallel optimizer IBM CPLEX.

The flexible socio spatial group queries

“…Shen et al [25] propose the multiple rally-point social spatial group query (MRGQ) that chooses a suitable rally point from the multiple points and the corresponding best group, which exhibits the minimized spatial distance between group members to the best rally point. The resulting group is of fixed size and satisfies the average minimum familiarity constraint.…”

“…Given such a network, socio spatial group queries [7,25,30,33] aim to find the 'best' group against a Point Of Interest (POI) where the users possess social tightness within the group and have spatial closeness to the POI. An example of such queries is to find a group of three members, who are located close to a particular restaurant and socially connected to at least one of the other members, so that the group is competent for a targeted advertisement of a "20% discount for a table of three" offer running at that restaurant.…”

“…Although existing works have contributions towards finding an important class of group queries, there exists several gaps with the real-life applications, particularly the following major limitations: (i) Impracticality of specifying group size: In each of the existing socio spatial group queries ( [7,25,30,33]), a single value as the size of the group (i.e., fixed size group) needs to be specified by the query issuer a priori. However, without prior knowledge of the social connections and users' locations, it is difficult to provide an exact and explicit size for the desired group.…”

“…For example, given multiple event locations of a festival, each resulting group can get advertisement for its nearest event. To the best of our knowledge, the only existing work that incorporates multiple POIs is [25]. However, their proposed algorithm can find only one group (i.e., k = 1) as the result.…”

“…As the existing studies [7,30,33] can only find the best group of only one fixed size for only one POI, their algorithm needs to be repeatedly reissued for each of the POIs and for each of the group sizes between the minimum and maximum values. Although Shen et al [25] allows multiple POIs as the query input, they can only find one single group (i.e., top-1 group) of a fixed size. Therefore, their approach cannot be directly applied to find the top-2 groups.…”