Research on Ecotourism Recommendation using Graph Embedding based Graph Neural Network

Le résumé fourni par la source

The increasing global emphasis on sustainable tourism has necessitated the development of intelligent ecotourism recommendation systems to enhance user experience and environmental conservation efforts. This study presents a novel approach that leverages Graph Neural Networks (GNN) to provide personalized and efficient ecotourism recommendations. The proposed model integrates various environmental and user preference parameters, including geographic data, biodiversity indices, climate conditions, and traveler interests, to generate optimized travel suggestions. Existing recommendation systems predominantly rely on collaborative filtering and content-based techniques, which frequently struggle to capture the complex relationships between different tourism destinations and user behaviors. Addressing these limitations, the GNN-based framework efficiently processes interconnected tourism data by modeling destinations as graph nodes and their relationships as weighted edges, thereby improving prediction accuracy and adaptability. The experimental evaluation, conducted on a curated ecotourism dataset, shows that the proposed model attains remarkable accuracy of 96.5%, significantly outperforming conventional approaches regarding recommendation precision, recall, and computational efficiency. By facilitating informed and sustainable travel decisions, the framework contributes to responsible ecotourism development, aligning with global sustainability objectives.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.