As digital transformation accelerates across industries, cybersecurity faces escalating challenges due to increasingly sophisticated cyber threats. This study explores the integration of artificial intelligence (AI) algorithms and big data analytics to enhance cybersecurity systems, focusing on addressing data integration and interpretability issues. Employing a descriptive-qualitative methodology, the research analyzes literature, case studies, and secondary data to evaluate the effectiveness of AI and big data in detecting and mitigating cyber threats. Key findings reveal that deep learning algorithms, such as artificial neural networks, achieved an accuracy of 93% in anomaly detection, outperforming traditional rule-based approaches by 18%. Additionally, big data platforms like Spark demonstrated superior efficiency, processing 500 GB of data in 35 seconds compared to Hadoop’s 60 seconds. However, the study identifies challenges related to the interpretability of AI models and the complexity of integrating diverse datasets, which impede real-time threat detection. Periodic updates to AI training datasets were found to improve detection accuracy by up to 15%, emphasizing the importance of adaptive learning models. This research contributes to the field by proposing strategies to enhance system resilience, including adopting Explainable AI (XAI) for transparency and advanced data integration techniques. The findings underscore the potential of AI and big data to revolutionize cybersecurity, offering organizations a proactive approach to combating evolving cyber threats. Future studies should focus on sector-specific implementations and optimizing response mechanisms for comprehensive security frameworks.
Read full abstract