Efficient Processing of Reachability Queries with Meetings

Abstract

The prevalence of location tracking systems has resulted in large volumes of spatiotemporal data generated every day. Addressing reachability queries on such datasets is important for a wide range of applications, such as security monitoring, surveillance, public health, epidemiology, social networks, etc. Given two objects OS, OT and a time interval I, a reachability query identifies whether information (or physical items etc.) could have transferred from OS to OT during I (typically indirectly through intermediaries). While traditional graph reachability queries have been studied extensively, little work exists on processing spatiotemporal reachability queries for large disk-resident trajectory datasets. Moreover, previous research assumed that information can be passed from one object to another instantaneously. However, in many applications such transfer takes time (i.e., a short conversation), thus forcing interacting objects to stay in contact for some time interval. This requirement makes the query processing even more challenging. In this paper, we introduce a novel problem, namely, spatiotemporal reachability queries with meetings and propose two algorithms, RICCmeetMin and RICCmeetMax. To prune the search space during query time, these algorithms precompute some reachability events: the shortest valid meetings (RICCmeetMin), and the longest possible meetings (RICCmeetMax) respectively. An extended experimental evaluation examines the efficiency and pruning characteristics of both algorithms over a variety of spatiotemporal reachability queries with meetings on large disk-resident datasets.