Large Language Models are Zero-Shot Next Location Predictors

TL;DR

This paper demonstrates that large language models can effectively predict future locations in human mobility scenarios without prior training on specific data, outperforming specialized models significantly.

Contribution

It introduces the novel application of LLMs as zero-shot next-location predictors, leveraging their generalization and reasoning capabilities for mobility prediction tasks.

Findings

LLMs achieved up to 36.2% accuracy in location prediction.

LLMs outperformed specialized mobility models by nearly 640%.

LLMs can provide explanations for their predictions regardless of size.

Abstract

Predicting the locations an individual will visit in the future is crucial for solving many societal issues like disease diffusion and reduction of pollution. However, next-location predictors require a significant amount of individual-level information that may be scarce or unavailable in some scenarios (e.g., cold-start). Large Language Models (LLMs) have shown good generalization and reasoning capabilities and are rich in geographical knowledge, allowing us to believe that these models can act as zero-shot next-location predictors. We tested more than 15 LLMs on three real-world mobility datasets and we found that LLMs can obtain accuracies up to 36.2%, a significant relative improvement of almost 640% when compared to other models specifically designed for human mobility. We also test for data contamination and explored the possibility of using LLMs as text-based explainers for next-location prediction, showing that, regardless of the model size, LLMs can explain their decision.