Jul 2026· 2026 International Conference on Intelligent and Sustainable AI Systems (ICOSAAS)· pp. 632-641· 0 citations· 7 references
Abstract
Large-scale DDoS attacks remain a serious threat to today's networked systems, which aim to make services unavailable by sending a massive amount of traffic. The traditional detection methods are mostly about attack categorization and are not that context-aware or actionable in providing support to security analysts. We propose SentinelsGuard AI, a self-learning DDoS classification and mitigation system that is built from a combination of machine learning, retrieval-augmented knowledge, and LLM-based reasoning in one platform. The proposed system is realized as a Flask-based web application, which processes the network flow statistics data, extracted from incoming traffic, and transforms the data to generate representative features of the network flows, before detecting the attacks using the XGBoost classifier. After classification, the framework accesses domain-specific attack intelligence from a structured Retrieval-Augmented Generation (RAG) knowledge base with profiles of multiple categories of DDoS attacks. This contextual information is used with LLM-enabled reasoning to derive comprehensible security explanations, determine the severity of the attack, and suggest the firewall mitigation rules that can be applied. The framework stores analysis data in a SQLite database for operational monitoring, and displays security events in an interactive dashboard that includes traffic visualization, attack logs, and threat summaries. The architecture proposed brings statistical learning together with contextual knowledge of cybersecurity and attempts to bridge the gap between automatic attack detection and explainable incident response. It integrates all the key features of a classification-based network security decision support platform into a single stream, enabling a classification-based approach to network security to become more interpretable and usable in reality: persistent logging, visualization, knowledge retrieval, reasoning, and system recommendation of mitigations.
AI-driven cyberattacks represent major challenges to detect by classic security frameworks, which makes it important to employ intelligent and autonomous mechanisms. In this paper, we present a data-driven analysis of network attack detection and reduction using machine learning, deep learning, and an Autonomous Defense Agent (ADA) for real-time threat detection and response. Experiments are done on two benchmark datasets CIC-IoT 2023 and UNSW-NB15 datasets, representing network intrusion scenarios. We addressed the imbalance in the dataset CIC-IoT 2023 with a 1:41 benign-to-attack skewed ratio by applying the Synthetic Minority Over-sampling Technique (SMOTE) to the training set. For classification progress, we used four classical ML classifiers (Decision Tree, Random Forest, XGBoost, and KNN), and also an LSTM model. Furthermore, we designed an Autonomous Defense Agent (ADA) for real-time intrusion detection. The highest F1 achieved for Random Forest is 99.39% on CIC-IoT 2023 and 88.67% on UNSW-NB15. For XGBoost, the same accuracy was achieved with 12 times faster training speed. This makes it the preferred ADA backbone. LSTM outperforms on structured IoT traffic with F1=98.82%, but on the other hand, fails on heterogeneous traffic with F1=43.29%. This shows how data temporal characteristics affect model selection. The ADA achieves 99.94% precision on CIC-IoT 2023 with sub-1ms average response latency. The results provide a principled model selection criterion that prefers the classical ML methods for heterogeneous traffic and LSTM for temporally structured IoT traffic, and also provides an ADA design to validate real benchmark datasets.
Marwah Yaseen· Al-Noor Journal of Engineeri...· 0 citations
Distributed Denial-of-Service (DDoS) attacks remain among the most disruptive threats to modern network infrastructure, with adversaries continually adapting their strategies to overwhelm cloud platforms, Internet-of-Things (IoT) deployments, and Software-Defined Network (SDN) environments. Traditional signature-based intrusion detection systems exhibit inherent inflexibility against novel attack vectors, motivating a shift toward intelligent, data-driven defense mechanisms. This paper presents an intelligent DDoS detection and mitigation framework that combines classical Machine Learning (ML) classifiers with Deep Learning (DL) architectures to achieve high-fidelity, low-latency attack identification across heterogeneous network topologies. Evaluated on the CICDDoS2019, NSL-KDD, and UNSW-NB15 benchmark datasets, the proposed hybrid framework incorporating XGBoost and a Bidirectional LSTM model achieves a classification accuracy of 99.31%, a precision of 99.18%, a recall of 99.27%, and an F1-score of 99.22%, outperforming standalone classifiers while sustaining sub-millisecond detection latency under realistic traffic loads. SDN-assisted rule insertion further reduces the mean mitigation response time to 8.4 ms. The results affirm the viability of deploying intelligent, explainable ML-based defense pipelines in production-grade network environments.
S. Singh, Alok Kumar· International Journal of Com...· 0 citations
Autonomous response has evolved into a timing-critical challenge rather than solely a matter of detection accuracy. In recent intrusions, the interval between initial access and the first lateral movement has been observed to be as short as 27 seconds, a window that precludes any human-in-the-loop workflow. This paper presents a closed-loop framework that detects and blocks attacks in software-defined networks without operator involvement, evaluating its performance against this stringent temporal constraint rather than relying exclusively on detection accuracy. An automated data pipeline collects IP flows and aggregates them into labeled training data, while a prevention module selects and trains candidate classifiers and issues blocking rules directly to the SDN controller. In a SYN flooding denial of service case study, the deployed K-Nearest Neighbors classifier achieved an F1 score of 96.7% and the cycle from flow availability to enforced block completed in 21 seconds, below the fastest breakout time reported to date.
The rapid advancement of digital communication and networking technologies has resulted in a significant increase in the frequency and sophistication of cyber threats, creating new challenges for securing modern computer networks. Traditional intrusion detection approaches mainly depend on signature-based techniques and predefined security rules, making them less effective against newly emerging and continuously evolving cyber attacks. To address these limitations, this study presents an Explainable Artificial Intelligence (XAI)-based cyber threat detection framework that combines Long Short-Term Memory (LSTM) and Autoencoder models for accurate and transparent threat detection. The LSTM model captures sequential network traffic patterns to identify temporal attack behaviours, whereas the Autoencoder detects anomalous activities by learning the characteristics of normal network traffic. The integration of Explainable AI enables users to understand prediction outcomes by highlighting the factors influencing each detection decision. Furthermore, the framework performs quantitative risk assessment, categorizes threats into multiple severity levels, and generates suitable mitigation recommendations through an interactive web-based dashboard. Experimental results demonstrate that the proposed framework provides reliable, interpretable, and effective cyber threat detection, thereby supporting security professionals in making timely and well-informed decisions.
Indu Asitha, M. N.· International Journal of Com...· 0 citations
Abstract - Rising volumes of sophisticated cyber attacks have rendered conventional signature-driven security tools insufficient for protecting modern digital assets. This study introduces a multi-layered intelligent detection system that unites supervised learning, unsupervised anomaly analysis, and ensemble decision strategies to identify network intrusions, malicious software activity, and stealthy advanced persistent threats in near real time. The pipeline begins with rigorous data cleansing and feature refinement that combines mutual-information ranking with principal-component analysis. Classification is performed by a soft-voting ensemble of Random Forest, gradient-boosted trees (XGBoost), and a compact deep neural network. Parallel anomaly scoring is obtained from Isolation Forest and a reconstruction-error autoencoder trained solely on benign flows. Evaluation on the NSL-KDD, CIC-IDS2017 and UNSW-NB15 collections yields detection accuracy reaching 98.7 %, precision of 97.9 %, recall of 98.4 % and an F1-score of 98.1 %, accompanied by a low rate of false alarms. SHAP-based explanations are attached to every prediction, giving security analysts transparent insight into model behaviour. Comparative benchmarks against recent published methods confirm that the proposed architecture delivers both higher accuracy and practical deployability inside enterprise security operations centres.
Key Words: network intrusion detection, ensemble machine learning, anomaly scoring, deep autoencoder, explainable artificial intelligence, cyber-attack classification, security operations.
Ameen Pasha.A· International Scientific Jou...· 0 citations
Controller Area Network (CAN) is the dominant in-vehicle bus, yet its broadcast design and absent authentication leave it exposed to injection and spoofing attacks. Existing deep-learning intrusion detection systems achieve strong accuracy but depend on ML inference frameworks incompatible with the resource budgets of production Electronic Control Units (ECUs). We present a systems-oriented pipeline that trains four neural architectures—Transformer, CNN, LSTM, and ANN—on a five-class labeled CAN dataset and distills each into a framework-free C++ decision tree via the Trustee protocol. We observe and address a systematic false-alarm collapse on Normal-class traffic: standard Trustee produces around 30% FPR-N because Normal frames constitute only 5.4% of training data. A class-balanced Trustee variant, which oversamples Normal by factor α=5, reduces FPR-N to 13.8±1.4% (three Trustee runs) while preserving 91.1±0.6% accuracy and 0.918±0.006 macro-F1 at 2.1% CPU on a Raspberry Pi 4, versus 22.6% for TF-Lite int8 and 59.4% for TensorFlow on the same board. We additionally report α and depth ablations, three-seed teacher statistics, and per-class recall for all models.Code: https://github.com/Amirmasoud-Pourmiri/Lightweight-Neural-Network-Distillation-for-Real-Time-CAN-Bus-Intrusion-Detection-in-Vehicles
Amirmasoud Pourmiri, Ali Eslami, Sergio A. Salinas Monroy· International Conference on...· 0 citations