On-line discovery of flock patterns in spatio-temporal data

Abstract

With the recent advancements and wide usage of location detection devices, large quantities of data are collected by GPS and cellular technologies in the form of trajectories. While most previous work on trajectory-based queries has concentrated on traditional range, nearest-neighbor and similarity queries, there is a increasing interest in queries that capture the "aggregate" behavior of trajectories as groups. Consider for example, finding groups of moving objects that move "together", i.e. within a predefined distance to each other, for a certain continuous period of time. Such queries typically arise in surveillance applications, e.g. identify groups of suspicious people, convoys of vehicles, flocks of animals, etc. In this paper we first show that the on-line flock discovery problem is polynomial and then propose a framework and several strategies to discover such patterns in streaming spatio-temporal data. Experiments with real and synthetic trajectorial datasets show that the proposed algorithms are efficient and scalable.

Visual mining of moving flock patterns in large spatio-temporal data sets using a frequent pattern approach

“…Vieira et al (2009) identified three groups of 'collective' patterns in moving object databases: clustering for moving objects, convoy queries and flock patterns. Moving clusters (Jensen et al 2007) and convoy queries (Jeung et al 2008) differ from flock patterns mainly because they do not necessarily contain the same objects during the lifetime of a cluster or convoy.…”

“…Indeed, Gudmundsson and van Kreveld (2006) have shown that the discovery of fixed flocks -defined as patterns where the same entities stay together during the entire interval -is an nondeterministic polynomial time-hard problem. Vieira et al (2009) were the first to present an exact solution for reporting flock patterns in polynomial time and also for those that can work effectively in real time. Their work revealed that a polynomial time solution can be found through identifying a discrete number of locations to place the centre of the flock disk.…”

“…The main objective is the generation of a final set of disks which cluster the number of trajectories in groups according to proximity criteria. This step still uses the parameter ", to define the diameter of the disks, and μ, for pruning procedures to reduce the number of valid disks, as proposed in Vieira et al (2009).…”

“…Of particular interest has been the development of approaches that can identify groups of moving objects whose members share a strong relationship by being present within a defined spatial region during a given time duration. Some examples of these kinds of approaches are moving cluster analyses (Kalnis et al 2005, Jensen et al 2007, convoy queries (Jeung et al 2008) and flock patterns (Gudmundsson and van Kreveld 2006, Benkert et al 2008, Vieira et al 2009). Vieira et al (2009) define moving flock patterns as groups of at least μ entities moving in the same direction while being close to one another during a given time interval δ (Figure 1).…”

GEOSO - A Geo-Social Model: From Real-World Co-occurrences to Social Connections

“…One area of related work includes a number of recent studies [20,21,22,23,24] that investigate similarity between objects' (e.g., people, cars) locations in time, represented as trajectories, for various reasons (e.g., to identify moving convoys, to recommend carpool partnerships). In particular, the similarity between trajectories can be used to infer social connections among people as shown by Li et al [3].…”