Integrating Predictive Analytics and Business Intelligence for Enterprise-Scale Cybersecurity Threat Detection in the United States

Cybersecurity attacks against enterprises in the United States have increased in terms of intensity, rate, and sophistication, thereby giving rise to a pressing need for the development of predictive defense tools capable of anticipating and detecting potential threats in advance. This paper works towards the development of an end-to-end business intelligence (BI) modeling platform utilizing the strengths of machine learning, anomaly identification, and real-time log analytics to formulate predictive models and provide recommendations to mitigate potential cyber security threats in the enterprise environment. Using a dataset represented by a collection of representative enterprise environment network log entries, user login activity, and system events, the platform makes judicious utilization of supervised and unsupervised learning approaches such as the application of the gradient boosting algorithm, random forests, and autoencoders to recognize precursor signs of hacking activities, malware diffusion, and insider attacks in the enterprise environment. The designed BI platform offers the unique capability of interpreting results from the predictive analytics components and converting them into a form of actional visual analytics to enable security teams to assess risks and automate the process of response to security breaches in the enterprise environment. The results validate the efficacy of the combined BI and machine learning platforms to enhance the threat detection rate by a maximum margin of 28% in comparison to the application of traditional rule-based systems while reducing mean time to detection and mean time to response times in the enterprise environment. The paper forecasts the augmentation of enterprise resilience and aligns with the strategic objectives of the United States regarding the objectives of the NIST AI Risk Management Framework and the National Cybersecurity Strategy. Future work will relate to large-scale implementation and the application of generative AI to enhance the security models designed to predict security breaches in the cloud environment.