Predicting Carbon Footprint of Tourism Activities Using Deep Neural Networks

Abstract

This study focuses on predicting the carbon footprint of tourism activities in Gansu Province using deep neural network technology. Based on recent tourism-related data, the research employs a Bidirectional Long Short-Term Memory (Bi-LSTM) network combined with an attention mechanism to construct the prediction model. Through systematic data preprocessing, feature engineering, and model optimization, the final model demonstrates superior predictive performance on the test set, significantly outperforming traditional methods and other machine learning models. The study also utilizes SHAP value analysis to reveal key factors influencing tourism carbon footprints. This research not only provides a reliable tool for predicting tourism carbon footprints but also offers important insights for formulating targeted carbon reduction strategies, contributing significantly to the sustainable development of the tourism industry.