Big data-assisted urban governance: The influence factors of citizens’ satisfaction on government hotline services

This study examines the modernization of Government Hotline Services (GHS) from legacy telephony to integrated multichannel platforms, highlighting the role of big data analytics in optimizing service efficiency and enhancing public trust. Using operational data from a Chinese GHS, two predictive models were developed: Our approach delivers a sentiment classifier with a 2% accuracy gain over the R-EKLT algorithm and a GRU-based satisfaction predictor that achieves 89.21% accuracy. A tripartite validation—citizen, government, and agent perspectives—confirmed reliability. Analysis revealed high-expectation factors (efficiency, quality, completion, and sentiment) and low-expectation factors (delays and repeated follow-ups) as key determinants of citizen satisfaction. Building on these insights, the study proposes a scalable data-driven governance framework to continuously predict satisfaction and dynamically adjust services in real time. This framework enhances service responsiveness, strengthens citizen-government trust, and improves urban governance efficiency, ultimately elevating the quality of life and tourism experiences in cities.