Fast discovery of multidimensional subsequences for robust trajectory classification

TL;DR

This paper introduces a fast method for discovering meaningful subtrajectories in large mobility datasets, optimizing the MASTERMovelets approach to improve trajectory classification efficiency and interpretability.

Contribution

It presents a novel, optimized algorithm for rapid subtrajectory discovery that enhances the existing MASTERMovelets method for trajectory classification.

Findings

Reduced search space accelerates subtrajectory discovery.

Improved classification accuracy with interpretable patterns.

Method outperforms previous approaches in speed and effectiveness.

Abstract

Trajectory classification tasks became more complex as large volumes of mobility data are being generated every day and enriched with new sources of information, such as social networks and IoT sensors. Fast classification algorithms are essential for discovering knowledge in trajectory data for real applications. In this work we propose a method for fast discovery of subtrajectories with the reduction of the search space and the optimization of the MASTERMovelets method, which has proven to be effective for discovering interpretable patterns in classification problems.