Sociodemographic data adds no clear value to predicting next locations
Who You Are Adds Nothing Detectable to Where You Go Next: Sociodemographic Conditioning in LLM Next-Location Prediction
Computers and Society
Summary
Predicting where people will go next using large language models is common, but it's unclear if personal details like age or income help. The authors tested this by linking movement data with demographics for 5,000 residents and found that adding these personal details made almost no difference in prediction accuracy. Interestingly, while past locations can reveal income, knowing income doesn’t improve next-move predictions. They also found that how possible next places are chosen to compare affects prediction results much more than demographics do.
What this means in practice
- •For urban planners: Design models for human movement predictions without relying on sociodemographic data to simplify data requirements.
- •For mobile app developers: Improve next-location prediction features by focusing on mobility history and candidate selection rather than demographic data.
Authors
Xin Wang, Paraic Carroll, Kerry Nice, Sachith Seneviratne, Li Zhang
Abstract
Large language models (LLMs) are increasingly used for individual next-location prediction, while sociodemographic conditioning is common in LLM-based travel simulation. Yet the incremental predictive value of sociodemographic attributes remains unclear. To directly test this contribution, sociodemographic records were linked with passively sensed mobility data from 5,000 Shenzhen residents to construct a closed-set benchmark in which models rank 100 candidate destinations. Each prediction instance is evaluated with and without age, gender, occupation and income, while holding mobility history, candidates and all other prompt content fixed. Results show that across four history lengths, the paired change in top-1 accuracy ranges from -0.8 to +0.5 percentage points, with no detectable gain from attributes. This result remains consistent when stay history is withheld, across alternative prediction times, in two additional LLMs and in a supervised reranker trained on the same benchmark. The null does not reflect a lack of model responsiveness to demographic information, as permuted attributes reduce LLM accuracy whereas correctly matched attributes do not improve it. A further asymmetry emerges in the reverse predictive direction, as pre-cut mobility trajectories recover income with an AUC of 0.708, while sociodemographic attributes contribute little to next-location prediction. Beyond demographic conditioning, candidate construction exerts a much larger influence on reported performance. Removing distance raises top-1 accuracy by 7.7 percentage points under proximity sampling but lowers it by 22.3 points under popularity sampling, with the reversal reproduced across all three LLMs. These results distinguish demographic association from incremental predictive usefulness and show that sampled next-location accuracy depends strongly on how candidate alternatives are constructed.