With the development of big data technology, network information security issues have received increasing attention. Government information systems are widely used in digital transformation, but it is difficult to avoid network information security problems. Therefore, based on big data technology, this paper constructs a government cloud security protection system using situational awareness technology to address the ever-growing network information security issues. Firstly, in the analysis of data collection and processing, this paper analyzes situational awareness, situational understanding, and situational prediction. Secondly, for accurate identification of security threats, this paper builds a security situational neural network training model based on the SA-GRU government cloud for quantitative evaluation of security situations. Thirdly, to ensure the secure operation of information systems, a security protection system is established based on the model, achieving information security protection for government cloud systems in terms of monitoring, early warning, and protection.