With the rapid development of modern Internet technology, network information, with its unique advantages of immediacy, interactivity, massive volume and multimedia, is widely used in enterprise operation systems as well as in social life, and the ensuing information security risks are highlighted. Based on this, this study mainly constructs an enterprise computer network information management and security policy protection model based on the improved SAE-K means++ algorithm, improves the convergence process of LSTM to enhance its global optimization-seeking capability, subsequently optimizes the model hyperparameters using the improved SAE-K means++ algorithm, and builds an enterprise computer network security posture prediction model The model was subsequently optimized using the improved SAE-K means++ algorithm, and an enterprise computer network security posture prediction model was built. This means that the system not only meets the national requirements for internal network information security protection capabilities, but also meets the market demand for IDC security.