User Satisfaction with Chinese Government Apps: Topic Mining and Sentiment Analysis of User Reviews

This study investigates user satisfaction with Chinese government apps, employing topic mining and sentiment analysis techniques to examine a dataset of 32,660 reviews from 10 province-level government apps. Using the Latent Dirichlet Allocation (LDA) topic model, we developed a customized framework for evaluating user satisfaction. We conducted a comparative analysis of eight machine learning (ML) models to identify the most effective approach for sentiment analysis and satisfaction quantification. The findings reveal that user satisfaction with these apps is moderately low, indicating challenges in mobile government service quality. Notably, significant regional disparities were observed, with higher satisfaction in central and eastern regions compared to northeastern and western regions. This reflects the influence of socioeconomic, geographic, and cultural factors on digital governance outcomes. Based on these insights, actionable recommendations for improving government app design and enhancing mobile government service quality are proposed.