AI-Driven Approaches toEnhancing Cybersecurity Posturing In Organizations
Kundan Kumar · African Journal of Biomedical Research · 2024
The threats in the cyber domain have become more complex and massive in nature and therefore organizations are struggling to have strong cybersecurity posture.The focus of this study is to examine the possibility of using Artificial Intelligence (AI) in refining cybersecurity by increasing threat detection, identifying anomalies and designing real-time response systems.Several approaches are explored in this study, including supervised algorithms such as Decision Tree and SVM, while the unsupervised strategies drawn include K-Means clustering and autoencoders.The study also shows that AI can enhance the detection accuracy, minimize false positives and increase the scalability of cybersecurity solutions.The results show that the proposed approach based on K-Means clustering increases the accuracy of anomaly detection by 30%, and autoencoders increase the rates of detection of complex attacks by 25% compared to traditional methods.Furthermore, we also discuss how deep RL can be used to learn a response to new threats, though the computational cost of this approach is currently prohibitive.The research also develops a novel cybersecurity model that combines both supervised and unsupervised models, which provides a stronger defense system than the individual approach.Of course, AI has significant potential benefits; complementary issues, including the requirement for supervised learning data and the high computational cost of most deep learning algorithms, still persist.This work highlights the need to create better, more interpretable, and efficient AI models to make such technologies available to every organization.The study's findings contribute to the body of knowledge in AI-based cybersecurity and offer practical recommendations for enhancing cybersecurity structures when counteracting advanced cyber threats.