Abstract:
The growing complexity, diversity, and sheer volume of cyberattacks have highlighted the inadequacies of traditional detection systems that depend on single-source data or isolated strategies. Today’s cyber threats often operate across various domains, exploiting multiple attack vectors and generating intricate patterns in disparate forms such as network flows, system logs, intrusion alerts, and endpoint telemetry. This level of sophistication demands advanced approaches to integrate and correlate heterogeneous cybersecurity data, enabling a unified, context-rich, and precise understanding of emerging threats. While significant progress has been made in intrusion detection and anomaly detection models, current methods frequently suffer from fragmented analyses, elevated false alarm rates, and computational overhead. To address these challenges, this research introduces a hybrid framework that employs mechanisms for integrating and correlating diverse data types to enhance the performance of detection systems. Leveraging the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology, the study systematically managed key processes such as data collection, pre-processing, model development, and evaluation. Three widely recognized cybersecurity datasets benchmark, CICIDS2018, NSL-KDD), and UNSW-NB15 were utilized independently to ensure robustness and cross-validation across a variety of attack scenarios. The hybrid model achieved an accuracy of 99.30%, 97.56% and 97.85 for CICIDS2018, NSL-KDD and UNSW-NB15 datasets, which is better than some of the existing methods. The proposed hybrid framework aimed to minimize false positives while improving detection accuracy for numerous attack categories. The model was implemented using Python mechanisms. This study ultimately underscores the effective integration and contextual correlation of diverse data sources can significantly enhance cyber defence capabilities while paving the way for adaptive, intelligent, and scalable intrusion detection systems.
Biography
Amadi Chukwuemeka is a cybersecurity researcher and technology-focused professional whose work lies at the intersection of cybersecurity, data analytics, and digital innovation. With a strong academic foundation in computer science and information security, his research focuses on advanced threat detection, heterogeneous data integration, and the application of intelligent analytics to complex digital environments. My academic work emphasizes on the integration and correlation of diverse cybersecurity data sources to improve situational awareness and support proactive cyber defence strategies. Through rigorous research methodologies and applied experimentation, I seek to bridge the gap between theoretical models and practical security solutions capable of addressing the evolving global cyber threat landscape. Beyond my technical research, I am deeply interested in the broader societal implications of emerging technologies guided by intellectual curiosity, discipline, and a strong commitment to impact, he aspires to contribute to global discussions on cybersecurity, artificial intelligence, and data-driven decision-making while helping shape secure and sustainable digital ecosystems.
