Generative Next POI Recommendation with Semantic ID

2025-06-02

AI Summary

Existing POI recommendation systems rely on arbitrary numeric IDs, hindering effective modeling of semantic relationships among locations. To address this, we propose a generative next-POI recommendation framework grounded in Semantic IDs (SIDs). Our approach introduces the first interpretable and generalizable semantic representation for POIs. We design a Residual Quantized Variational Autoencoder (RQ-VAE) to discretize the POI semantic space, coupled with a diversity loss to ensure uniform distribution of SIDs within that space. Furthermore, the framework jointly incorporates user behavioral patterns, collaborative filtering signals, geographic features, and fine-tuned large language model capabilities. Evaluated on three benchmark datasets, our method consistently outperforms state-of-the-art approaches, achieving up to a 16% improvement in recommendation accuracy. It demonstrates superior generalization across sparse and cold-start scenarios while preserving full interpretability through human-readable semantic IDs.

Technology Category

• Knowledge Representation and Reasoning: Preferences
• Data Mining & Knowledge Management: Recommender Systems
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Point-of-interest (POI) recommendation systems aim to predict the next destinations of user based on their preferences and historical check-ins. Existing generative POI recommendation methods usually employ random numeric IDs for POIs, limiting the ability to model semantic relationships between similar locations. In this paper, we propose Generative Next POI Recommendation with Semantic ID (GNPR-SID), an LLM-based POI recommendation model with a novel semantic POI ID (SID) representation method that enhances the semantic understanding of POI modeling. There are two key components in our GNPR-SID: (1) a Semantic ID Construction module that generates semantically rich POI IDs based on semantic and collaborative features, and (2) a Generative POI Recommendation module that fine-tunes LLMs to predict the next POI using these semantic IDs. By incorporating user interaction patterns and POI semantic features into the semantic ID generation, our method improves the recommendation accuracy and generalization of the model. To construct semantically related SIDs, we propose a POI quantization method based on residual quantized variational autoencoder, which maps POIs into a discrete semantic space. We also propose a diversity loss to ensure that SIDs are uniformly distributed across the semantic space. Extensive experiments on three benchmark datasets demonstrate that GNPR-SID substantially outperforms state-of-the-art methods, achieving up to 16% improvement in recommendation accuracy.

Problem

• Enhances semantic understanding in POI recommendation systems
• Improves recommendation accuracy using semantic ID representation
• Generates semantically rich POI IDs for better modeling

Innovation

• LLM-based POI recommendation with semantic IDs
• Semantic ID Construction using collaborative features
• Residual quantized variational autoencoder for POI quantization