Exploring the ways of the Internet in shaping low-carbon behavior by using PLS-SEM and machine learning algorithms

Abstract The Internet plays a pivotal role in tackling global climate challenges, particularly in facilitating a transition toward low-carbon behavior. However, the mechanisms by which the Internet shapes low-carbon behavior remain inadequately understood. This study investigates the influence of the Internet on low-carbon behavior through three primary pathways: information dissemination, technology adoption, and trust. This study uses data collected from an online survey of 1308 respondents conducted in China. By integrating PLS-SEM with machine learning, specifically Artificial Neural Networks (ANN) and Generalized Additive Models (GAM), this study offers a comprehensive method for understanding the complex relationships between variables. The findings reveal that the Internet fosters low-carbon behavior by enhancing low-carbon knowledge, awareness, climate change risk perception, and social influence. Internet-based low-carbon behavior applications’ perceived usefulness and ease of use significantly encourage low-carbon behavior, while trust in online information and applications acts as a critical indirect driver. The main contribution of this study is the development of a novel conceptual framework that explains how low-carbon behavior is shaped in the digital age. The results provide a theoretical foundation for policymakers to design strategies that leverage the Internet for advocacy, education, and technology to advance sustainable development.