Enhancing government text big data recognition and management via reinforcement learning optimization

With the rapid development of big data and machine learning technologies, there is an increasing demand for text data processing and management in the field of government affairs. In order to improve the processing speed and efficiency of governmental texts in smart city construction, the study proposes an improved inverse reinforcement learning framework. The method constructs an algorithmic framework based on reinforcement learning to obtain governmental text data through automated crawlers and identify and process them. The experimental results show that the proposed model outperforms the comparison model in terms of accuracy, precision, recall and F1 value, and the proposed model achieves a precision rate of 94.6%, a recall rate of 92.4%, and an F1 value of 0.94. In addition, the model demonstrates good real-time processing capability with a training time of 4.3 hours and an inference time of 24ms/sample. The experimental results show that the proposed model is able to improve the recognition and processing efficiency in governmental text. The research provides theoretical support for improving the level of urban governance intelligence and optimizing the process of government services.