Sustainable tourism development through eco-security evaluation: insights from DPSIR-SBM and machine learning models
摘要
Tourism eco-security is the foundation of sustainable tourism development. This study aims to reveal the dynamic evolution, influencing factors and future trends of tourism eco-security, in order to address core problems such as ill-defined eco-security thresholds and imperfect early warning mechanisms in sustainable tourism development. According to the DPSIR-SBM model, this research constructs a tourism eco-security evaluation index system and analyzes the spatio-temporal evolution characteristics. The quantitative regression method is used to analyze the influencing mechanism of tourism eco-security. The CNN-LSTM model is used to reasonably predict the future pattern of tourism eco-security. The findings are as follows: The average tourism eco-security level shows a fluctuating upward trend, while the trend of polarization is weakening, and the difference in tourism eco-security shows a fluctuating downward trend. There are significant positive or negative effects of different influencing factors of tourism eco-security, but the magnitude of coefficients of each variable at different quartiles varies significantly different. From 2023 to 2030, tourism eco-security will show fluctuating changes; the overall trend will be an upward spiral. However, the evolution of tourism eco-security is accompanied by the phenomenon of “alternating ups and downs,” and the potential risks to tourism eco-security persist. This study provides theoretical support and practical guidance for the management of regional eco-security by constructing a scientific evaluation system and prediction model, and addresses deficiencies in the analysis of multi-scale spatial and temporal characteristics and nonlinear relationships in the existing studies.