Search results

The objective of this study was to analyse the influencing factors of citizens' dissatisfaction with government services during the COVID-19 pandemic to help government departments identify problems in the service process and possible countermeasures.

The authors first used cosine interesting pattern mining (CIPM) to analyse citizens' complaints in different periods of the pandemic. Second, the potential evaluation indices of customer satisfaction were extracted from the hotline business system through a hypothesis analysis and modelled using multiple regression analysis. During the index transformation and standardization process, a machine-learning algorithm of clustering and emotion analysis was adopted. Finally, the authors used the random forest algorithm to evaluate the importance of the indicators and obtain the indicators more important to citizen satisfaction.

The authors found that the complaint topic, appeal time, urgency of citizens' complaints, citizens' emotions, level of detail in the case record, and processing timeliness and efficiency significantly influenced citizens' satisfaction. When the government addresses complaints in a more standardized and efficient manner, citizens are more satisfied.

During the pandemic, government departments should be more patient with citizens, increase the speed of the case circulation and shorten the processing period of appeals. Staff should record appeals in a more standardized manner, highlighting themes and prioritizing urgent cases to appease citizens and relieve their anxiety.