Consumers' attitudes toward low-carbon consumption based on a computational model: Evidence from China

The COP26 conference emphasized the "shift from commitment to action", and the Chinese government has actively issued a series of policies related to the low-carbon economy, which have impacted Chinese consumers' attitudes toward low-carbon consumption. However, one of the current challenges is the perspectives used to understand attitudes toward low carbon consumption vary, and the research methodology relies heavily on traditional methods with a time lag. This paper presents a computational model that combines Sentiment Classification (SC), the Latent Dirichlet Allocation (LDA) model, and VOSviewer clustering to show persuasive topics from consumers. We illustrate the validity of the model and apply it to new energy vehicle posts from social media platforms (e.g., Zhihu) between January 2022 and March 2022. Finally, we observe 14 clusters with sentiment and find that consumers' attitudes are polarized, positive stakeholders consider the long-term development and macroscopic aspects, while negative stakeholders believe that there are real problems with new energy vehicles, especially batteries and charging, and then we provide some suggestions to the government, manufacturers, and investors.