Explore the vast range of titles available.

The development of smart network infrastructure of the Internet of Things (IoT) faces the immense threat of sophisticated Distributed Denial-of-Services (DDoS) security attacks. The existing network security solutions of enterprise networks are significantly expensive and unscalable for IoT. The integration of recently developed Software Defined Networking (SDN) reduces a significant amount of computational overhead for IoT network devices and enables additional security measurements. At the prelude stage of SDN-enabled IoT network infrastructure, the sampling based security approach currently results in low accuracy and low DDoS attack detection. In this paper, we propose an Adaptive Machine Learning based SDN-enabled Distributed Denial-of-Services attacks Detection and Mitigation (AMLSDM) framework. The proposed AMLSDM framework develops an SDN-enabled security mechanism for IoT devices with the support of an adaptive machine learning classification model to achieve the successful detection and mitigation of DDoS attacks. The proposed framework utilizes machine learning algorithms in an adaptive multilayered feed-forwarding scheme to successfully detect the DDoS attacks by examining the static features of the inspected network traffic. In the proposed adaptive multilayered feed-forwarding framework, the first layer utilizes Support Vector Machine (SVM), Naive Bayes (NB), Random Forest (RF), k-Nearest Neighbor (kNN), and Logistic Regression (LR) classifiers to build a model for detecting DDoS attacks from the training and testing environment-specific datasets. The output of the first layer passes to an Ensemble Voting (EV) algorithm, which accumulates the performance of the first layer classifiers. In the third layer, the adaptive frameworks measures the real-time live network traffic to detect the DDoS attacks in the network traffic. The proposed framework utilizes a remote SDN controller to mitigate the detected DDoS attacks over Open Flow (OF) switches and reconfigures the network resources for legitimate network hosts. The experimental results show the better performance of the proposed framework as compared to existing state-of-the art solutions in terms of higher accuracy of DDoS detection and low false alarm rate.

The research examines how RL and DRL models can be used to enhance the prediction of maintenance needs in the IIoT setting. The purpose is to assess the accuracy, precision, recall, F1 score and the AUC-ROC of adaptive models against non-adaptive models. It is clear from the results that adaptive models outperform traditional models in fault prediction, providing better accuracy and more accurate predictions. Furthermore, adaptive models can handle changes in the environment and the equipment better than other models. Moreover, when these models are used with edge and cloud computing, they make sure that decisions are applied quickly and that the models can be easily integrated into industrial systems. The research also demonstrates that adaptive machine learning models can improve the accuracy of the model and reduce both false positive and false negative cases. When compared to non-adaptive baselines, adaptive models increased recall by up to 11.2% points and precision by up to 10.2% points. The Adaptive Ensemble performed best overall (93.4% accuracy, 95.2% AUC-ROC). Experimental assessment reveals consistent and statistically significant enhancements in performance for adaptive models across all criteria. The Adaptive Ensemble attains superior performance, achieving 93.4% accuracy and 95.2% AUC-ROC. In comparison to the most robust non-adaptive baseline (Random Forest), it enhances memory by 8.5% points, precision by 7.8% points, and F1-score by 8.2% points. In comparison to SVM, recall increases by 11.2% points and precision by 10.2% points, signifying significant decreases in undetected faults and false positives.The study provides information about how adaptive learning can be used in IIoT-based PdM systems and offers advice to industries that want to make their PdM systems more reliable, effective and cost-efficient.

In this manuscript, we develop a multi-party framework tailored for multiple data contributors seeking machine learning insights from combined data sources. Grounded in statistical learning principles, we introduce the Multi-Key Homomorphic Encryption Logistic Regression (MK-HELR) algorithm, designed to execute logistic regression on encrypted multi-party data. Given that models built on aggregated datasets often demonstrate superior generalization capabilities, our approach offers data contributors the collective strength of shared data while ensuring their original data remains private due to encryption. Apart from facilitating logistic regression on combined encrypted data from diverse sources, this algorithm creates a collaborative learning environment with dynamic membership. Notably, it can seamlessly incorporate new participants during the learning process, addressing the key limitation of prior methods that demanded a predetermined number of contributors to be set before the learning process begins. This flexibility is crucial in real-world scenarios, accommodating varying data contribution timelines and unanticipated fluctuations in participant numbers, due to additions and departures. Using the AI4I public predictive maintenance dataset, we demonstrate the MK-HELR algorithm, setting the stage for further research in secure, dynamic, and collaborative multi-party learning scenarios.

Automation of machine learning model development is increasingly becoming an established research area. While automated model selection and automated data pre-processing have been studied in depth, there is, however, a gap concerning automated model adaptation strategies when multiple strategies are available. Manually developing an adaptation strategy can be time consuming and costly. In this paper we address this issue by proposing the use of flexible adaptive mechanism deployment for automated development of adaptation strategies. Experimental results after using the proposed strategies with five adaptive algorithms on 36 datasets confirm their viability. These strategies achieve better or comparable performance to the custom adaptation strategies and the repeated deployment of any single adaptive mechanism.

Phishing detection models degrade quickly due to drift, adversarial evasion, and fairness issues. Existing MLOps platforms mainly automate deployment and monitoring. Prior works have examined SHAP-based monitoring, retraining, or fairness audits separately, but lack an integrated theory of resilience for adversarial environments. We introduce the Hybrid MLOps Framework (HAMF), a system designed to embed resilience and ethical governance into the lifecycle of phishing detection models. HAMF is ‘hybrid’ because it unifies proactive and reactive adaptation, combining automation with stakeholder oversight, and embedding resilience with ethical governance. HAMF treats resilience as an integrated lifecycle property, designed to simultaneously preserve model accuracy, fairness, and stakeholder trust amidst concept drift. Methodologically, HAMF implements this through a hybrid control cycle. This cycle fuses four key capabilities: SHAP-guided feature replacement, event-driven retraining, fairness-triggered audits, and structured human feedback. Unlike conventional pipelines where these functions are isolated, HAMF ensures their interdependence as first-class triggers. Empirical evaluations on large-scale phishing streams demonstrate HAMF’s superior performance. The framework detects drift within 18 seconds, restores F1 scores above 0.99 post-attack, reduces subgroup disparities by over 60%, and scales to over 2,300 requests per second with sub-50ms latency. These results validate HAMF’s design, demonstrating that embedding resilience and ethical alignment into the MLOps lifecycle is both effective and scalable.

Adaptive machine learning has increasing importance due to its ability to classify a data stream and handle the changes in the data distribution. Various resources, such as wearable sensors and medical devices, can generate a data stream with an imbalanced distribution of classes. Many popular oversampling techniques have been designed for imbalanced batch data rather than a continuous stream. This work proposes a self-adjusting window to improve the adaptive classification of an imbalanced data stream based on minimizing cluster distortion. It includes two models; the first chooses only the previous data instances that preserve the coherence of the current chunk’s samples. The second model relaxes the strict filter by excluding the examples of the last chunk. Both models include generating synthetic points for oversampling rather than the actual data points. The evaluation of the proposed models using the Siena EEG dataset showed their ability to improve the performance of several adaptive classifiers. The best results have been obtained using Adaptive Random Forest in which Sensitivity reached 96.83% and Precision reached 99.96%.

The number of bits required in phase shifters (PS) in hybrid precoding (HP) has a significant impact on sum-rate, spectral efficiency (SE), and energy efficiency (EE). The space and cost constraints of a realistic massive multiple-input multiple-output (MIMO) system limit the number of antennas at the base station (BS), limiting the throughput gain promised by theoretical analysis. This paper demonstrates the effectiveness of employing an intelligent reflecting surface (IRS) to enhance efficiency, reduce costs, and conserve energy. Particularly, an IRS consists of an extensive number of reflecting elements, wherein every individual element has a distinct phase shift. Adjusting each phase shift and then jointly optimizing the source precoder at BS and selecting the optimal phase-shift values at IRS will allow us to modify the direction of signal propagation. Additionally, we can improve sum-rate, EE, and SE performance. Furthermore, we proposed an energy-efficient HP at BS in which the analog component is implemented using a low-resolution PS rather than a high-resolution PS. Our analysis reveals that the performance gets better as the number of bits increases. We formulate the problem of jointly optimizing the source precoder at BS and the reflection coefficient at IRS to improve the system performance. However, because of the non-convexity and high complexity of the formulated problem. Inspired by the cross-entropy (CE) optimization technique used in machine learning, we proposed an adaptive cross-entropy (ACE) 1-3-bit PS-based optimization HP approach for this new architecture. Moreover, our analysis of energy consumption revealed that increasing the low-resolution bits can significantly reduce power consumption while also improving performance parameters such as SE, EE, and sum-rate. The simulation results are presented to validate the proposed algorithm, which highlights the IRS efficiency gains to boost sum-rate, SE, and EE compared to previously reported methods.

In order to make design decisions, engineers may seek to identify regions of the design domain that are acceptable in a computationally efficient manner. A design is typically considered acceptable if its reliability with respect to parametric uncertainty exceeds the designer’s desired level of confidence. Despite major advancements in reliability estimation and in design classification via decision boundary estimation, the current literature still lacks a design classification strategy that incorporates parametric uncertainty and desired design confidence. To address this gap, this works offers a novel interpretation of the acceptance region by defining the decision boundary as the hypersurface which isolates the designs that exceed a user-defined level of confidence given parametric uncertainty. This work addresses the construction of this novel decision boundary using computationally efficient algorithms that were developed for reliability analysis and decision boundary estimation. The proposed approach is verified on two physical examples from structural and thermal analysis using Support Vector Machines and Efficient Global Optimization-based contour estimation.

This paper introduces an adaptive interval fuzzy modeling method using participatory learning and interval-valued stream data. The model is a collection of fuzzy functional rules in which the rule base structure and the parameters of the rules evolve simultaneously as data are input. The evolving nature of the method allows continuous model adaptation using the stream interval input data. The method employs participatory learning to cluster the interval input data recursively, constructs a fuzzy rule for each cluster, uses the weighted recursive least squares to update the parameters of the rule consequent intervals, and returns an interval-valued output. The method is evaluated using actual data to model and forecast the daily lowest and highest prices of the four most traded cryptocurrencies, BitCoin, Ethereum, XRP, and LiteCoin. The performance of the adaptive interval fuzzy modeling is compared with the adaptive neuro-fuzzy inference system, long short-term memory neural network, autoregressive integrated moving average, exponential smoothing state model, and the naïve random walk methods. Results show that the suggested interval fuzzy model outperforms all these methods in predicting prices in the digital coin market, especially when considering directional accuracy measure.

Modern network infrastructure faces unprecedented challenges in intelligent path selection due to exponential traffic growth and dynamic conditions. Current approaches suffer from either inflexibility (traditional protocols) or computational overheads (neural networks). This study presents HyperPath-SVM, a novel enhanced Support Vector Machine (SVM) framework addressing these limitations through three key innovations. First, Dynamic Discriminative Weight Evolution (DDWE) enables continuous weight adaptation through closed-form mathematical updates. Second, Temporal Graph Convolution Kernel (TGCK) incorporates network topology dynamics into kernel computations. Third, quantum-inspired optimisation implemented on classical hardware achieves faster convergence using principles from quantum annealing. Our evaluation of 127 million routing decisions collected over 8 months from real network datasets demonstrates exceptional performance: 96.5% path selection accuracy, 1.8 ms inference time, and 98 MB memory footprint. The framework maintains 94% accuracy during single-link and cascading network failures while providing complete interpretability for operational deployment. Production simulations indicate 31% latency reduction and 28% throughput improvement over the traditional protocols. This work establishes enhanced SVMs as superior alternatives to neural networks for real-time network intelligence, combining computational efficiency with adaptive learning capabilities.