Paper Title
Fortifying Cyber Defenses: Evaluating Modern IDS Paradigms Across Diverse Attack Landscapes

Abstract
Intrusion Detection Systems (IDS) play a vital role in safeguarding networks from unauthorized access and malicious activities. As cybersecurity threats continue to evolve, there is a growing need for detection methods that are both accurate and adaptable. This study presents a comparison of four modern IDS approaches—Federated Learning, Adversarial Resilience, Cross-Domain Learning, and Neuro-Symbolic methods—applied across three well-established datasets: KDD99, UNSW-NB15, and NF-UQ-NIDS. A review of prior research helped guide the selection of these techniques, each of which addresses a specific challenge in intrusion detection, such as privacy, robustness, generalization, or interpretability. The methodology involved standardized preprocessing, consistent evaluation metrics, and adversarial testing. Results showed that while all methods performed well in detecting threats, each had unique strengths depending on the application context. The study concludes that using a combination of these strategies can improve the effectiveness of IDS in complex and real-world environments.