Stop@: A framework for scalable and noise-resistant stop-move segmentation of large datasets of trajectories in outdoor and indoor spaces

Abstract

Capturing the mobility behavior of moving entities from their traces is a prominent theme in mobility data science. Stop@ supports behavior analysis by providing a generic framework for the mining of stop-move patterns in spatial trajectories across animal and human mobility scenarios. The framework is built around a stop detection method, successfully used in diverse applications in animal ecology. The method has been recently validated against accurate ground truth stops collected in a museum, proving to be effective and robust, also for the study of human mobility. Stop@ provides a rich set of functionalities to facilitate the stop-move analysis, including the parallel processing of large datasets of trajectories collected outdoor and indoor.

Keywords

• Human and animal mobility
• Trajectories
• Stop-move detection