Search Results (261)

This paper presents an attack detection, response, and recommendation framework designed to protect the integrity and operational continuity of IoT-based critical infrastructure, specifically focusing on an energy use case. With the growing deployment of IoT-enabled smart meters in energy systems, ensuring data integrity is essential. The proposed framework monitors smart meter data in real time, identifying deviations that may indicate data tampering or device malfunctions. The system comprises two main components: an attack detection and prediction module based on machine learning (ML) models and a response and adaptation module that recommends countermeasures. The detection module employs a forecasting model using a long short-term memory (LSTM) architecture, followed by a dense layer to predict future readings. It also integrates a statistical thresholding technique based on Tukey’s fences to detect abnormal deviations. The system was evaluated on real smart meter data in a testbed environment. It achieved accurate forecasting (MAPE < 2% in most cases) and successfully flagged injected anomalies with a low false positive rate, an effective result given the lightweight, unsupervised, and real-time nature of the approach. These findings confirm the framework’s applicability in resource-constrained energy systems requiring real-time cyberattack detection and mitigation. Full article

Human memory is inherently susceptible to errors, including the formation of false memories—instances where individuals mistakenly recall information they were never exposed to. While prior research has largely focused on neural activity associated with false memory, the structural brain correlates of this phenomenon remain relatively unexplored. This study bridges that gap by investigating gray matter structure as it relates to individual differences in false memory performance. Using publicly available magnetic resonance imaging datasets, we analyzed cortical thickness (CT) in neural regions implicated in memory processes. To assess false memory, we applied signal detection theory, which provides a robust framework for differentiating between true and false memory. Our findings reveal that increased CT in the parietal lobe and middle occipital gyrus correlates with greater susceptibility to false memories, highlighting its role in integrating and manipulating memory information. Conversely, CT in the middle frontal gyrus and occipital pole was associated with enhanced accuracy in memory recall, emphasizing its importance in perceptual processing and encoding true memories. These results provide novel insights into the structural basis of memory errors and offer a foundation for future investigations into the neural underpinnings of memory reliability. Full article

Advanced Persistent Threat (APT) attacks pose a serious challenge to traditional detection methods. These methods often suffer from high false-alarm rates and limited accuracy due to the multi-stage and covert nature of APT attacks. In this paper, we propose TSE-APT, a time-series ensemble model that addresses these two limitations. It combines multiple machine-learning models, such as Random Forest (RF), Multi-Layer Perceptron (MLP), and Bidirectional Long Short-Term Memory Network (BiLSTM) models, to dynamically capture correlations between multiple stages of the attack process based on time-series features. It discovers hidden features through the integration of multiple machine-learning models to significantly improve the accuracy and robustness of APT detection. First, we extract a collection of dynamic time-series features such as traffic mean, flow duration, and flag frequency. We fuse them with static contextual features, including the port service matrix and protocol type distribution, to effectively capture the multi-stage behaviors of APT attacks. Then, we utilize an ensemble-learning model with a dynamic weight-allocation mechanism using a self-attention network to adaptively adjust the sub-model contribution. The experiments showed that using time-series feature fusion significantly enhanced the detection performance. The RF, MLP, and BiLSTM models achieved 96.7% accuracy, considerably enhancing recall and the false positive rate. The adaptive mechanism optimizes the model’s performance and reduces false-alarm rates. This study provides an analytical method for APT attack detection, considering both temporal dynamics and context static characteristics, and provides new ideas for security protection in complex networks. Full article

To address the high false alarm rate of intrusion detection systems based on distributed acoustic sensing (DAS) for power cables in complex underground environments, an innovative GRT-Transformer multimodal deep learning model is proposed. The core of this model lies in its distinctive three-branch parallel collaborative architecture: two branches employ Gramian Angular Summation Field (GASF) and Recursive Pattern (RP) algorithms to convert one-dimensional intrusion waveforms into two-dimensional images, thereby capturing rich spatial patterns and dynamic characteristics and the third branch utilizes a Gated Recurrent Unit (GRU) algorithm to directly focus on the temporal evolution features of the waveform; additionally, a Transformer component is integrated to capture the overall trend and global dependencies of the signals. Ultimately, the terminal employs a Bidirectional Long Short-Term Memory (BiLSTM) network to perform a deep fusion of the multidimensional features extracted from the three branches, enabling a comprehensive understanding of the bidirectional temporal dependencies within the data. Experimental validation demonstrates that the GRT-Transformer achieves an average recognition accuracy of 97.3% across three typical intrusion events—illegal tapping, mechanical operations, and vehicle passage—significantly reducing false alarms, surpassing traditional methods, and exhibiting strong practical potential in complex real-world scenarios. Full article

Objective: This proof-of-concept study aimed to evaluate and compare the clinical performance of two AI language models (ChatGPT-3.5 and OpenAI o3) in answering clinical pharmacy multiple-choice questions (MCQs), benchmarked against responses from specialist clinical pharmacists in Jordan, including academic preceptors and hospital-based clinicians. Methods: A total of 60 clinical pharmacy MCQs were developed based on current guidelines across four therapeutic areas: cardiovascular, endocrine, infectious, and respiratory diseases. Each item was reviewed by academic and clinical experts and then pilot-tested with five pharmacists to determine clarity and difficulty. Two ChatGPT models—GPT-3.5 and OpenAI o3—were tested using a standardized prompt for each MCQ, entered in separate sessions to avoid memory retention. Their answers were classified as true/false positives or negatives and retested after two weeks to assess reproducibility. Simultaneously, 25 licensed pharmacists (primarily from one academic institution and several hospitals in Amman) completed the same MCQs using validated references (excluding AI tools). Accuracy, sensitivity, specificity, and Cohen’s Kappa were used to compare AI and human performance, with statistical analysis conducted using appropriate tests at a significance level of p ≤ 0.05. Results: OpenAI o3 achieved the highest accuracy (83.3%), sensitivity (90.0%), and specificity (70.0%), outperforming GPT-3.5 (70.0%, 77.5%, 55.0%) and pharmacists (69.7%, 77.0%, 55.0%). AI performance declined significantly with increasing question difficulty. OpenAI o3 showed the highest accuracy in the cardiovascular domain (93.3%), while GPT-3.5 performed best in infectious diseases (80.0%). Reproducibility was higher for GPT-3.5 (81.6%, κ = 0.556) than OpenAI o3 (76.7%, κ = 0.364). Over two test rounds, GPT-3.5’s accuracy remained stable, whereas OpenAI o3’s accuracy decreased from 83.3% to 70.0%, indicating some variability. Conclusions: OpenAI o3 shows strong promise as a clinical decision-support tool in pharmacy, especially for low- to moderate-difficulty questions. However, inconsistencies in reproducibility and limitations in complex cases highlight the importance of cautious, supervised integration alongside human expertise. Full article

This study presents a novel spatio-temporal detection framework for identifying False Data Injection (FDI) attacks in DC microgrid systems from the perspective of cyber–physical symmetry. While modern DC microgrids benefit from increasingly sophisticated cyber–physical symmetry network integration, this interconnected architecture simultaneously introduces significant cybersecurity vulnerabilities. Notably, FDI attacks can effectively bypass conventional Chi-square detector-based protection mechanisms through malicious manipulation of communication layer data. To address this critical security challenge, we propose a hybrid deep learning framework that synergistically combines: Convolutional Neural Networks (CNN) for robust spatial feature extraction from power system measurements; Long Short-Term Memory (LSTM) networks for capturing complex temporal dependencies; and an attention mechanism that dynamically weights the most discriminative features. The framework operates through a hierarchical feature extraction process: First-level spatial analysis identifies local measurement patterns; second-level temporal analysis detects sequential anomalies; attention-based feature refinement focuses on the most attack-relevant signatures. Comprehensive simulation studies demonstrate the superior performance of our CNN-LSTM-Attention framework compared to conventional detection approaches (CNN-SVM and MLP), with significant improvements across all key metrics. Namely, the accuracy, precision, F1-score, and recall could be improved by at least 7.17%, 6.59%, 2.72% and 6.55%. Full article

Fake news is false or misleading information that looks like real news and spreads through traditional and social media. It has a big impact on our social lives, especially in politics. In Pakistan, where Urdu is the main language, finding fake news in Urdu is difficult because there are not many effective systems for this. This study aims to solve this problem by creating a detailed process and training models using machine learning, deep learning, and large language models (LLMs). The research uses methods that look at the features of documents and classes to detect fake news in Urdu. Different models were tested, including machine learning models like Naïve Bayes and Support Vector Machine (SVM), as well as deep learning models like Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM), which used embedding techniques. The study also used advanced models like BERT and GPT to improve the detection process. These models were first evaluated on the Bend-the-Truth dataset, where CNN achieved an F1 score of 72%, Naïve Bayes scored 78%, and the BERT Transformer achieved the highest F1 score of 79% on Bend the Truth dataset. To further validate the approach, the models were tested on a more diverse dataset, Ax-to-Grind, where both SVM and LSTM achieved an F1 score of 89%, while BERT outperformed them with an F1 score of 93%. Full article

With the expansion of oil and gas exploration and development to complex oil and gas resource areas such as deep and ultra-deep formation onshore and offshore, the kick is one of the high drilling risks, and timely and accurate early kick detection is increasingly important. Based on the kick generation mechanism, kick characterization parameters are preliminarily selected. According to the characteristics of the data and previous research progress, Random Forest (RF), Support Vector Machine (SVM), Feedforward Neural Network (FNN), and Long Short-term Memory Neural Network (LSTM) are established using experimental data from Memorial University of Newfoundland. The test results show that the accuracy of the SVM-linear model was 0.968, and the missing alarm and the false alarm rate only was 0.06 and 0.11. Additionally, through the analysis of the kick response time, the lag time of the SVM-linear model was 1.3 s, and the comprehensive equivalent time was 23.13 s, which showed the best performance. The different effects of the model after data transformation are analyzed, the mechanism of the best effect of the SVM model is analyzed, and the changes in the effect of other models including RF are further revealed. The proposed early-warning model warns in advance in historical well logging data, which is expected to provide a fast, efficient, and accurate gas kick warning model for drilling sites. Full article

The exponential growth of Distributed Denial-of-Service (DDoS) attacks in high-speed networks presents significant real-time detection and mitigation challenges. The existing detection frameworks are categorized into flow-based and packet-based detection approaches. Flow-based approaches usually suffer from high latency and controller overhead in high-volume traffic. In contrast, packet-based approaches are prone to high false-positive rates and limited attack classification, resulting in delayed mitigation responses. To address these limitations, we propose a real-time DDoS detection architecture that combines hardware-accelerated statistical preprocessing with GPU-accelerated deep learning models. The raw packet header information is transformed into multivariate time series data to enable classification of complex traffic patterns using Temporal Convolutional Networks (TCN), Long Short-Term Memory (LSTM) networks, and Transformer architectures. We evaluated the proposed system using experiments conducted under low to high-volume background traffic to validate each model’s robustness and adaptability in a real-time network environment. The experiments are conducted across different time window lengths to determine the trade-offs between detection accuracy and latency. The results show that larger observation windows improve detection accuracy using TCN and LSTM models and consistently outperform the Transformer in high-volume scenarios. Regarding model latency, TCN and Transformer exhibit constant latency across all window sizes. We also used SHAP (Shapley Additive exPlanations) analysis to identify the most discriminative traffic features, enhancing model interpretability and supporting feature selection for computational efficiency. Among the experimented models, TCN achieves the most balance between detection performance and latency, making it an applicable model for the proposed architecture. These findings validate the feasibility of the proposed architecture and support its potential as a real-time DDoS detection application in a realistic high-speed network. Full article

In today’s business environment, emails are essential across all sectors, including finance and academia. There are two main types of emails: ham (legitimate) and spam (unsolicited). Spam wastes consumers’ time and resources and poses risks to sensitive data, with volumes doubling daily. Current spam identification methods, such as Blocklist approaches and content-based techniques, have limitations, highlighting the need for more effective solutions. These constraints call for detailed and more accurate approaches, such as machine learning (ML) and deep learning (DL), for realistic detection of new scams. Emphasis has since been placed on the possibility that ML and DL technologies are present in detecting email spam. In this work, we have succeeded in developing a hybrid deep learning model, where Long Short-Term Memory (LSTM) and the Gated Recurrent Unit (GRU) are applied distinctly to identify spam email. Despite the fact that the other models have been applied independently (CNNs, LSTM, GRU, or ensemble machine learning classifier) in previous studies, the given research has provided a contribution to the existing body of literature since it has managed to combine the advantage of LSTM in capturing the long-term dependency and the effectiveness of GRU in terms of computational efficiency. In this hybridization, we have addressed key issues such as the vanishing gradient problem and outrageous resource consumption that are usually encountered in applying standalone deep learning. Moreover, our proposed model is superior regarding the detection accuracy (90%) and AUC (98.99%). Though Transformer-based models are significantly lighter and can be used in real-time applications, they require extensive computation resources. The proposed work presents a substantive and scalable foundation to spam detection that is technically and practically dissimilar to the familiar approaches due to the powerful preprocessing steps, including particular stop-word removal, TF-IDF vectorization, and model testing on large, real-world size dataset (Enron-Spam). Additionally, delays in the feature comparison technique within the model minimize false positives and false negatives. Full article

Underwater images often exhibit characteristics such as low contrast, blurred and small targets, object clustering, and considerable variations in object morphology. Traditional detection methods tend to be susceptible to omission and false positives under these circumstances. Furthermore, owing to the constrained memory and limited computing power of underwater robots, there is a significant demand for lightweight models in underwater object detection tasks. Therefore, we propose an enhanced lightweight YOLOv10n-based model, BSE-YOLO. Firstly, we replace the original neck with an improved Bidirectional Feature Pyramid Network (Bi-FPN) to reduce parameters. Secondly, we propose a Multi-Scale Attention Synergy Module (MASM) to enhance the model’s perception of difficult features and make it focus on the important regions. Finally, we integrate Efficient Multi-Scale Attention (EMA) into the backbone and neck to improve feature extraction and fusion. The experiment results demonstrate that the proposed BSE-YOLO reaches 83.7% mAP@0.5 on URPC2020 and 83.9% mAP@0.5 on DUO, with the parameters reducing 2.47 M. Compared to the baseline model YOLOv10n, our BSE-YOLO improves mAP@0.5 by 2.2% and 3.0%, respectively, while reducing the number of parameters by approximately 0.2 M. The BSE-YOLO achieves a good balance between accuracy and lightweight, providing an effective solution for underwater object detection. Full article

Learning discriminative features between abnormal and normal instances is crucial for video anomaly detection within the multiple instance learning framework. Existing methods primarily focus on instances with the highest anomaly scores, neglecting the identification and differentiation of hard samples, leading to misjudgments and high false alarm rates. To address these challenges, we propose a dual triplet contrastive loss strategy. This approach employs dual memory units to extract four key feature categories: hard samples, negative samples, positive samples, and anchor samples. Contrastive loss is utilized to constrain the distance between hard samples and other samples, enabling accurate identification of hard samples and enhancing the discriminative ability of hard samples and abnormal features. Additionally, a multi-scale feature perception module is designed to capture feature information at different levels, while an adaptive global–local feature fusion module constructs complementary feature enhancement through feature fusion. Experimental results demonstrate the effectiveness of our method, achieving AUC scores of 87.16% on the UCF-Crime dataset and AP scores of 83.47% on the XD-Violence dataset. Full article

The widespread adoption of the Internet of Things (IoT) raises significant concerns regarding security and energy efficiency, particularly for low-resource devices. To address these IoT issues, we propose a cross-layer IoT architecture employing machine learning (ML) models and lightweight cryptography. Our proposed solution is based on role-based access control (RBAC), ensuring secure authentication in large-scale IoT deployments while preventing unauthorized access attempts. We integrate layer-specific ML models, such as long short-term memory networks for temporal anomaly detection and decision trees for application-layer validation, along with adaptive speck encryption for the dynamic adjustment of cryptographic overheads. We then introduce a granular RBAC system that incorporates energy-aware policies. The novelty of this work is the proposal of a cross-layer IoT architecture that harmonizes ML-driven security with energy-efficient operations. The performance of the proposed cross-layer system is evaluated by extensive simulations. The results obtained show that the proposed system can reduce false positives up to 32% and enhance system security by preventing unauthorized access up to 95%. We also achieve 30% reduction in power consumption using the proposed lightweight Speck encryption method compared to the traditional advanced encryption standard (AES). By leveraging convolutional neural networks and ML, our approach significantly enhances IoT security and energy efficiency in practical scenarios such as smart cities, homes, and schools. Full article

Landslides are among the most destructive geological hazards, necessitating precise landslide susceptibility mapping (LSM) for effective risk management. This study focuses on the northeastern region of Leshan City and investigates the influence of various non-landslide sampling strategies and machine learning (ML) models on LSM performance. Ten landslide conditioning factors, selected by SHAP analysis, and six models were utilized: Convolutional neural networks (CNNs), Long Short-Term Memory (LSTM), CNN-LSTM, CNN-LSTM with an attention mechanism (AM), Random Forest (RF), and eXtreme Gradient Boosting combined with Logistic Regression (XGBoost-LR). Three non-landslide sampling strategies were designed, with the certainty factor-based approach demonstrating superior performance by effectively capturing geological and physical characteristics, applying spatial constraints to exclude high-risk zones, and achieving improved mean squared error (MSE) and area under the curve (AUC) values. The results reveal that traditional ML models struggle with complex nonlinear relationships and imbalanced datasets, often leading to high false positive rates. In contrast, deep learning (DL) models—particularly CNN-LSTM-AM—achieved the best performance, with an AUC of 0.9044 and enhanced balance in accuracy, precision, recall, and Kappa. These improvements are attributed to the model’s ability to extract static spatial features (via CNNs), capture dynamic temporal patterns (via LSTM), and emphasize key features through the attention mechanism. This integrated architecture enhances the capacity to process heterogeneous data and extract landslide-relevant features. Overall, optimizing non-landslide sampling strategies, incorporating comprehensive geophysical information, enforcing spatial constraints, and enhancing feature extraction capabilities are essential for improving the accuracy and reliability of LSM. Full article

The existing financial payment system has inherent credit card fraud problems that must be solved with strong and effective solutions. In this research, a combined deep learning model that incorporates a convolutional neural network (CNN), long-short-term memory (LSTM), and fully connected output layer is proposed to enhance the accuracy of fraud detection, particularly in addressing the class imbalance problem. A CNN is used for spatial features, LSTM for sequential information, and a fully connected output layer for final decision-making. Furthermore, SMOTE is used to balance the data and hyperparameter tuning is utilized to achieve the best model performance. In the case of hyperparameter tuning, the detection rate is greatly enhanced. High accuracy metrics are obtained by the proposed CNN-LSTM (CLST) model, with a recall of 83%, precision of 70%, F1-score of 76% for fraudulent transactions, and ROC-AUC of 0.9733. The proposed model’s performance is enhanced by hyperparameter optimization to a recall of 99%, precision of 83%, F1-score of 91% for fraudulent cases, and ROC-AUC of 0.9995, representing almost perfect fraud detection along with a low false negative rate. These results demonstrate that optimization of hyperparameters and layers is an effective way to enhance the performance of hybrid deep learning models for financial fraud detection. While prior studies have investigated hybrid structures, this study is distinguished by its introduction of an optimized of CNN and LSTM integration within a unified layer architecture. Full article