SST: A Simplified Swin Transformer-based Model for Taxi Destination Prediction based on Existing Trajectory

TL;DR

This paper introduces SST, a simplified Swin Transformer model tailored for taxi destination prediction from trajectory data, achieving higher accuracy than existing methods by leveraging the consecutive nature of trajectory sequences.

Contribution

The paper proposes a simplified Swin Transformer architecture that omits the shifted window mechanism, specifically designed for trajectory data, and demonstrates its superior performance on real-world datasets.

Findings

SST outperforms state-of-the-art methods in accuracy.

The simplified model reduces complexity while maintaining high performance.

Trajectory data's sequential nature benefits from the tailored Transformer design.

Abstract

Accurately predicting the destination of taxi trajectories can have various benefits for intelligent location-based services. One potential method to accomplish this prediction is by converting the taxi trajectory into a two-dimensional grid and using computer vision techniques. While the Swin Transformer is an innovative computer vision architecture with demonstrated success in vision downstream tasks, it is not commonly used to solve real-world trajectory problems. In this paper, we propose a simplified Swin Transformer (SST) structure that does not use the shifted window idea in the traditional Swin Transformer, as trajectory data is consecutive in nature. Our comprehensive experiments, based on real trajectory data, demonstrate that SST can achieve higher accuracy compared to state-of-the-art methods.