Wetland Ecotourism Development Using Deep Learning and Grey Clustering Algorithm from the Perspective of Sustainable...

Area of Science:

• Ecotourism research
• Environmental management
• Data science applications

Background:

• Sustainable development of wetland ecotourism is crucial for China.
• Accurate prediction of passenger flow and water quality is needed for effective management.
• Zhangye Heihe wetland ecotourism spot serves as the case study.

Purpose of the Study:

• To promote sustainable development of wetland ecotourism in China.
• To plan passenger flow considering different tourism periods.
• To evaluate water quality in relation to tourism seasons.

Main Methods:

• Developed two single wetland ecotourism Demand Prediction Models (DPMs) using optimized Fuzzy Clustering Algorithm (FCA), grey theory, and Markov Chain.
• Proposed a hybrid DPM combining the two single models: optimized fuzzy grey clustering algorithm.
• Developed a Water Quality Evaluation (WQE) model using Deep Learning Backpropagation Neural Network (DL BPNN).

Main Results:

• The hybrid DPM achieved the highest accuracy with a Mean Absolute Percentage Error (MAPE) of 1.25% and Root Mean Square Error (RMSE) of 0.2532.
• Single models showed higher errors (MAPE: 11.67%, 1.45%; RMSE: 0.2526, 0.1652).
• Water quality is significantly better in the wet season compared to dry and flat seasons.

Conclusions:

• The hybrid model offers superior accuracy and stability for predicting wetland ecotourism passenger flow.
• Water quality is a critical factor, with wet seasons showing better conditions.
• Integrating environmental factors like water quality and passenger flow is essential for ecotourism strategies.