Classical Feature Embeddings Help in BERT-Based Human Mobility Prediction
This addresses the need for better mobility predictions for applications like disaster relief and city planning, but it is incremental as it builds on existing BERT-based models with added features.
The paper tackled the problem of human mobility forecasting by enriching a BERT-based model with temporal descriptors and POI embeddings to capture semantic context, resulting in significant accuracy improvements: GEO-BLEU scores increased from 0.34 to 0.75 for single-city and from 0.34 to 0.56 for multi-city prediction.
Human mobility forecasting is crucial for disaster relief, city planning, and public health. However, existing models either only model location sequences or include time information merely as auxiliary input, thereby failing to leverage the rich semantic context provided by points of interest (POIs). To address this, we enrich a BERT-based mobility model with derived temporal descriptors and POI embeddings to better capture the semantics underlying human movement. We propose STaBERT (Semantic-Temporal aware BERT), which integrates both POI and temporal information at each location to construct a unified, semantically enriched representation of mobility. Experimental results show that STaBERT significantly improves prediction accuracy: for single-city prediction, the GEO-BLEU score improved from 0.34 to 0.75; for multi-city prediction, from 0.34 to 0.56.