Spet Algorithm: Stop and Proximity Episodes in Trajectories

Abstract

ABSTRACT: In this paper, we propose the SPET (Stop and Proximity Episodes in Trajectories) algorithm to identify stop and proximity episodes in trajectories. A trajectory is the record of the evolution of the position of an object that is moving in space during a given time interval in order to achieve a goal. A stop is an episode of a trajectory during which the object remained continuously inside a point of interest (POI) a minimum time (specified by the business analysts) and a proximity is an episode of a trajectory during which the object remained continuously near a POI a minimum time. These episodes may help to understand the behavior of moving objects in several domains. For example, proximities episodes can help in advertising, where agents can identify appropriate spots in order to try to increase the visibility of certains POIs. In order to prove the feasibility and expediency of our proposal, we conduct a series of experiments with real vehicle trajectories, in neighborhoods (the POIs) of Rio de Janeiro. Our results reveal information that can be useful for traffic analysis about the density of visits and proximities of vehicles to these neighborhoods.