BERT4Loc: BERT for Location -- POI Recommender System

TL;DR

This paper introduces BERT4Loc, a BERT-based model that enhances POI recommendations by integrating user preferences and location data, outperforming existing sequential models on benchmark datasets.

Contribution

The study presents a novel BERT-based approach for personalized POI recommendation that effectively combines location and user preference information.

Findings

BERT4Loc outperforms state-of-the-art sequential models on benchmark datasets.

The model effectively integrates location data with user preferences.

Experimental results demonstrate improved recommendation quality.

Abstract

Recommending points of interest (POIs) is a challenging task that requires extracting comprehensive location data from location-based social media platforms. To provide effective location-based recommendations, it's important to analyze users' historical behavior and preferences. In this study, we present a sophisticated location-aware recommendation system that uses Bidirectional Encoder Representations from Transformers (BERT) to offer personalized location-based suggestions. Our model combines location information and user preferences to provide more relevant recommendations compared to models that predict the next POI in a sequence. Our experiments on two benchmark dataset show that our BERT-based model outperforms various state-of-the-art sequential models. Moreover, we see the effectiveness of the proposed model for quality through additional experiments.