Study on Tourism Development Using CRITIC Method for Tourist Satisfaction

This paper presents a novel approach for evaluating tourist satisfaction and developing optimized strategies by integrating the CRITIC method, deep learning with Multilayer Perceptron (MLP), and Genetic Algorithms (GA). The CRITIC method was employed to calculate the weights of various satisfaction indicators, such as accommodation quality and local services, based on their variance and correlation. These weights informed the MLP model, which accurately predicted tourist satisfaction with a mean absolute error (MAE) of 0.12 and a root mean square error (RMSE) of 0.18. Using the GA, the study identified optimal strategy combinations that improved satisfaction scores by up to 15% compared to baseline strategies. The integration of CRITIC, MLP, and GA provided a comprehensive framework that not only enhanced the accuracy of satisfaction evaluations but also optimized development strategies effectively. This approach demonstrated significant improvements in tourist satisfaction across various regions, with notable results including an increase in satisfaction scores from 4.0 to 4.3 in specific areas. The paper highlights the efficacy of combining these advanced methods to address complex evaluation and optimization challenges in tourism. Key contributions include the development of a robust, data-driven methodology for strategy optimization and the provision of actionable insights into satisfaction drivers, offering a valuable tool for tourism managers and policymakers. Future research directions involve expanding the methodology to different tourism sectors and incorporating real-time data for even more dynamic evaluations.