Optimizing Ecotourism in North Taihu Lake, Wuxi City, China: Integrating Back Propagation Neural Networks and Ant-Colony Algorithm for Sustainable Route Planning

Abstract

Urbanization's rapid pace has sparked a growing interest in nature-focused travel experiences, highlighting the growing importance of ecotourism. This study presents an innovative algorithm for ecotourism route planning, focusing on aligning tourists with attractions to enhance growth and appeal. The research utilizes ecological attractions in the Taihu Lake scenic area as an experimental dataset, incorporating historical travel data to examine the relationship between user characteristics and ecotourism attractions. Backpropagation neural networks and one-hot encoding are employed to predict visitor experiences. At the same time, a new ecotourism route design method combining deep learning and an ant colony algorithm based on average distance is applied to formulate an optimal ecotourism route. Results indicate Yuan Tou Zhu and Ling Shan as the top recommended destinations, with the optimal path identified as 1, 2, 3, 6, 4, 5, 7. This suggests that considering individual tourist preferences significantly elevates visitor satisfaction in ecotourism route planning, and it reveals the positive impact of aligning tourist attributes with attraction features. The findings underscore the importance of integrating user preferences into ecotourism planning strategies. Prioritizing personalized tourist experiences significantly enhances the effectiveness of ecotourism route planning initiatives. The research contributes a comprehensive framework for revitalizing ecotourism in the digital age, recommending the prioritization of individual tourist inclinations and attraction compatibility. Furthermore, adopting deep learning techniques and one-hot encoding is suggested to enhance the accuracy and efficacy of ecotourism planning.