Аңдатпа
This study proposes an intelligent discovery and potential assessment framework for China’s tourism and recreation resources based on big data and knowledge graph technologies. By integrating multi-source heterogeneous tourism data, a tourism knowledge graph containing 23,487 entity nodes and 156,329 relationship edges was constructed. An improved Node2Vec algorithm was employed for graph embedding and representation learning, combined with the Analytic Hierarchy Process (AHP) and entropy weight method to build a comprehensive evaluation model, achieving multi-dimensional quantitative assessment of tourism resource potential. Experimental results show that this method achieves 92.3% accuracy in resource discovery, with a correlation coefficient of 0.867 between evaluation results and actual tourism development indicators. The research identified 138 high-potential but underdeveloped tourism resource points, providing data-driven decision support for regional tourism planning and validating the effectiveness and practical value of knowledge graph technology in tourism resource management.