Diabetes is increasing commonly in people’s daily life and represents an extraordinary threat to human well-being.Machine Learning(ML)in the healthcare industry has recently made headlines.Several ML models are devel...Diabetes is increasing commonly in people’s daily life and represents an extraordinary threat to human well-being.Machine Learning(ML)in the healthcare industry has recently made headlines.Several ML models are developed around different datasets for diabetic prediction.It is essential for ML models to predict diabetes accurately.Highly informative features of the dataset are vital to determine the capability factors of the model in the prediction of diabetes.Feature engineering(FE)is the way of taking forward in yielding highly informative features.Pima Indian Diabetes Dataset(PIDD)is used in this work,and the impact of informative features in ML models is experimented with and analyzed for the prediction of diabetes.Missing values(MV)and the effect of the imputation process in the data distribution of each feature are analyzed.Permutation importance and partial dependence are carried out extensively and the results revealed that Glucose(GLUC),Body Mass Index(BMI),and Insulin(INS)are highly informative features.Derived features are obtained for BMI and INS to add more information with its raw form.The ensemble classifier with an ensemble of AdaBoost(AB)and XGBoost(XB)is considered for the impact analysis of the proposed FE approach.The ensemble model performs well for the inclusion of derived features provided the high Diagnostics Odds Ratio(DOR)of 117.694.This shows a high margin of 8.2%when compared with the ensemble model with no derived features(DOR=96.306)included in the experiment.The inclusion of derived features with the FE approach of the current state-of-the-art made the ensemble model performs well with Sensitivity(0.793),Specificity(0.945),DOR(79.517),and False Omission Rate(0.090)which further improves the state-of-the-art results.展开更多
This research work proposes a new stack-based generalization ensemble model to forecast the number of incidences of conjunctivitis disease.In addition to forecasting the occurrences of conjunctivitis incidences,the pr...This research work proposes a new stack-based generalization ensemble model to forecast the number of incidences of conjunctivitis disease.In addition to forecasting the occurrences of conjunctivitis incidences,the proposed model also improves performance by using the ensemble model.Weekly rate of acute Conjunctivitis per 1000 for Hong Kong is collected for the duration of the first week of January 2010 to the last week of December 2019.Pre-processing techniques such as imputation of missing values and logarithmic transformation are applied to pre-process the data sets.A stacked generalization ensemble model based on Auto-ARIMA(Autoregressive Integrated Moving Average),NNAR(Neural Network Autoregression),ETS(Exponential Smoothing),HW(Holt Winter)is proposed and applied on the dataset.Predictive analysis is conducted on the collected dataset of conjunctivitis disease,and further compared for different performance measures.The result shows that the RMSE(Root Mean Square Error),MAE(Mean Absolute Error),MAPE(Mean Absolute Percentage Error),ACF1(Auto Correlation Function)of the proposed ensemble is decreased significantly.Considering the RMSE,for instance,error values are reduced by 39.23%,9.13%,20.42%,and 17.13%in comparison to Auto-ARIMA,NAR,ETS,and HW model respectively.This research concludes that the accuracy of the forecasting of diseases can be significantly increased by applying the proposed stack generalization ensemble model as it minimizes the prediction error and hence provides better prediction trends as compared to Auto-ARIMA,NAR,ETS,and HW model applied discretely.展开更多
Distribution transformers play a vital role in power distribution systems,and their reliable operation is crucial for grid stability.This study presents a simulation-based framework for active fault diagnosis and earl...Distribution transformers play a vital role in power distribution systems,and their reliable operation is crucial for grid stability.This study presents a simulation-based framework for active fault diagnosis and early warning of distribution transformers,integrating Sample Ensemble Learning(SEL)with a Self-Optimizing Support Vector Machine(SO-SVM).The SEL technique enhances data diversity and mitigates class imbalance,while SO-SVM adaptively tunes its hyperparameters to improve classification accuracy.A comprehensive transformer model was developed in MATLAB/Simulink to simulate diverse fault scenarios,including inter-turn winding faults,core saturation,and thermal aging.Feature vectors were extracted from voltage,current,and temperature measurements to train and validate the proposed hybrid model.Quantitative analysis shows that the SEL–SO-SVM framework achieves a classification accuracy of 97.8%,a precision of 96.5%,and an F1-score of 97.2%.Beyond classification,the model effectively identified incipient faults,providing an early warning lead time of up to 2.5 s before significant deviations in operational parameters.This predictive capability underscores its potential for preventing catastrophic transformer failures and enabling timely maintenance actions.The proposed approach demonstrates strong applicability for enhancing the reliability and operational safety of distribution transformers in simulated environments,offering a promising foundation for future real-time and field-level implementations.展开更多
Background:Stomach cancer(SC)is one of the most lethal malignancies worldwide due to late-stage diagnosis and limited treatment.The transcriptomic,epigenomic,and proteomic,etc.,omics datasets generated by high-through...Background:Stomach cancer(SC)is one of the most lethal malignancies worldwide due to late-stage diagnosis and limited treatment.The transcriptomic,epigenomic,and proteomic,etc.,omics datasets generated by high-throughput sequencing technology have become prominent in biomedical research,and they reveal molecular aspects of cancer diagnosis and therapy.Despite the development of advanced sequencing technology,the presence of high-dimensionality in multi-omics data makes it challenging to interpret the data.Methods:In this study,we introduce RankXLAN,an explainable ensemble-based multi-omics framework that integrates feature selection(FS),ensemble learning,bioinformatics,and in-silico validation for robust biomarker detection,potential therapeutic drug-repurposing candidates’identification,and classification of SC.To enhance the interpretability of the model,we incorporated explainable artificial intelligence(SHapley Additive exPlanations analysis),as well as accuracy,precision,F1-score,recall,cross-validation,specificity,likelihood ratio(LR)+,LR−,and Youden index results.Results:The experimental results showed that the top four FS algorithms achieved improved results when applied to the ensemble learning classification model.The proposed ensemble model produced an area under the curve(AUC)score of 0.994 for gene expression,0.97 for methylation,and 0.96 for miRNA expression data.Through the integration of bioinformatics and ML approach of the transcriptomic and epigenomic multi-omics dataset,we identified potential marker genes,namely,UBE2D2,HPCAL4,IGHA1,DPT,and FN3K.In-silico molecular docking revealed a strong binding affinity between ANKRD13C and the FDA-approved drug Everolimus(binding affinity−10.1 kcal/mol),identifying ANKRD13C as a potential therapeutic drug-repurposing target for SC.Conclusion:The proposed framework RankXLAN outperforms other existing frameworks for serum biomarker identification,therapeutic target identification,and SC classification with multi-omics datasets.展开更多
Modern intrusion detection systems(MIDS)face persistent challenges in coping with the rapid evolution of cyber threats,high-volume network traffic,and imbalanced datasets.Traditional models often lack the robustness a...Modern intrusion detection systems(MIDS)face persistent challenges in coping with the rapid evolution of cyber threats,high-volume network traffic,and imbalanced datasets.Traditional models often lack the robustness and explainability required to detect novel and sophisticated attacks effectively.This study introduces an advanced,explainable machine learning framework for multi-class IDS using the KDD99 and IDS datasets,which reflects real-world network behavior through a blend of normal and diverse attack classes.The methodology begins with sophisticated data preprocessing,incorporating both RobustScaler and QuantileTransformer to address outliers and skewed feature distributions,ensuring standardized and model-ready inputs.Critical dimensionality reduction is achieved via the Harris Hawks Optimization(HHO)algorithm—a nature-inspired metaheuristic modeled on hawks’hunting strategies.HHO efficiently identifies the most informative features by optimizing a fitness function based on classification performance.Following feature selection,the SMOTE is applied to the training data to resolve class imbalance by synthetically augmenting underrepresented attack types.The stacked architecture is then employed,combining the strengths of XGBoost,SVM,and RF as base learners.This layered approach improves prediction robustness and generalization by balancing bias and variance across diverse classifiers.The model was evaluated using standard classification metrics:precision,recall,F1-score,and overall accuracy.The best overall performance was recorded with an accuracy of 99.44%for UNSW-NB15,demonstrating the model’s effectiveness.After balancing,the model demonstrated a clear improvement in detecting the attacks.We tested the model on four datasets to show the effectiveness of the proposed approach and performed the ablation study to check the effect of each parameter.Also,the proposed model is computationaly efficient.To support transparency and trust in decision-making,explainable AI(XAI)techniques are incorporated that provides both global and local insight into feature contributions,and offers intuitive visualizations for individual predictions.This makes it suitable for practical deployment in cybersecurity environments that demand both precision and accountability.展开更多
The growing use of Portable Document Format(PDF)files across various sectors such as education,government,and business has inadvertently turned them into a major target for cyberattacks.Cybercriminals take advantage o...The growing use of Portable Document Format(PDF)files across various sectors such as education,government,and business has inadvertently turned them into a major target for cyberattacks.Cybercriminals take advantage of the inherent flexibility and layered structure ofPDFs to inject malicious content,often employing advanced obfuscation techniques to evade detection by traditional signature-based security systems.These conventional methods are no longer adequate,especially against sophisticated threats like zero-day exploits and polymorphic malware.In response to these challenges,this study introduces a machine learning-based detection framework specifically designed to combat such threats.Central to the proposed solution is a stacked ensemble learning model that combines the strengths of four high-performing classifiers:Random Forest(RF),Extreme Gradient Boosting(XGB),LightGBM(LGBM),and CatBoost(CB).These models operate in parallel as base learners,each capturing different aspects of the data.Their outputs are then refined by a Gradient Boosting Classifier(GBC),which serves as a meta-learner to enhance prediction accuracy.To ensure the model remains both efficient and effective,Principal Component Analysis(PCA)is applied to reduce feature dimensionality while preserving critical information necessary for malware classification.The model is trained and validated using the CIC-Evasive PDFMalware2022 dataset,which includes a wide range of both malicious and benign PDF samples.The results demonstrate that the framework achieves impressive performance,with 97.10% accuracy and a 97.39% F1-score,surpassing several existing techniques.To enhance trust and interpretability,the system incorporates Local Interpretable Model-agnostic Explanations(LIME),which provides user-friendly insights into the rationale behind each prediction.This research emphasizes how the integration of ensemble learning,feature reduction,and explainable AI can lead to a practical and scalable solution for detecting complex PDF-based threats.The proposed framework lays the foundation for the next generation of intelligent,resilient cybersecurity systems that can address ever-evolving attack strategies.展开更多
Background:Early detection of harmful brain activity in critically ill patients using electroencephalography(EEG)is vital for timely and effective clinical intervention.Automating EEG analysis with deep learning techn...Background:Early detection of harmful brain activity in critically ill patients using electroencephalography(EEG)is vital for timely and effective clinical intervention.Automating EEG analysis with deep learning techniques holds significant promise for enhancing diagnostic efficiency and accuracy.Methods:We implemented EfficientNetB2,which leverages convolutional neural networks with a novel Temporal Squeeze-and-Excitation module to capture temporal EEG features,and WaveNet,a sequential model designed to effectively model temporal dependencies in EEG data using dilated causal convolutions and temporal self-attention.Both models were trained and evaluated using a publicly available EEG dataset,with performance assessed via 4-fold cross-validation and a step-wise learning rate reduction strategy.Results:Our results demonstrate a significant reduction in training loss from 0.6459 to 0.3055 and validation loss from 0.9602 to 0.5719 over six epochs.Consistent improvements were observed across cross-validation folds,highlighting the robustness of the models.Additionally,ensemble learning of the two architectures further enhanced classification performance.Conclusion:This comparative analysis sheds light on the strengths and limitations of EfficientNetB2 and WaveNet for automated harmful brain activity detection in EEG signals.The findings contribute to the advancement of reliable and efficient deep learning models,paving the way for their clinical application in managing critically ill patients.展开更多
We propose a novel cooling protocol within a triple-Laguerre-Gaussian cavity optomechanical system,which is designed to suppress the thermal vibrations of a rotating mirror to reach its quantum ground state.The system...We propose a novel cooling protocol within a triple-Laguerre-Gaussian cavity optomechanical system,which is designed to suppress the thermal vibrations of a rotating mirror to reach its quantum ground state.The system incorporates two auxiliary cavities and an atomic ensemble coupled to a Laguerre-Gaussian rotational cavity.By carefully selecting system parameters,the cooling process of the rotating mirror is significantly enhanced,while the heating process is effectively suppressed,enabling efficient ground-state cooling even in the unresolved sideband regime.Compared to previous works,our scheme reduces the stringent restrictions on auxiliary systems,making it more experimentally feasible under broader parameter conditions.These findings provide a robust approach for achieving ground-state cooling in mechanical resonators.展开更多
Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional ...Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional domain adaptation methods assume a single source domain,making them less suitable for modern deep learning settings that rely on diverse and large-scale datasets.To address this limitation,recent research has focused on Multi-Source Domain Adaptation(MSDA),which aims to learn effectively from multiple source domains.In this paper,we propose Efficient Domain Transition for Multi-source(EDTM),a novel and efficient framework designed to tackle two major challenges in existing MSDA approaches:(1)integrating knowledge across different source domains and(2)aligning label distributions between source and target domains.EDTM leverages an ensemble-based classifier expert mechanism to enhance the contribution of source domains that are more similar to the target domain.To further stabilize the learning process and improve performance,we incorporate imitation learning into the training of the target model.In addition,Maximum Classifier Discrepancy(MCD)is employed to align class-wise label distributions between the source and target domains.Experiments were conducted using Digits-Five,one of the most representative benchmark datasets for MSDA.The results show that EDTM consistently outperforms existing methods in terms of average classification accuracy.Notably,EDTM achieved significantly higher performance on target domains such as Modified National Institute of Standards and Technolog with blended background images(MNIST-M)and Street View House Numbers(SVHN)datasets,demonstrating enhanced generalization compared to baseline approaches.Furthermore,an ablation study analyzing the contribution of each loss component validated the effectiveness of the framework,highlighting the importance of each module in achieving optimal performance.展开更多
Processes supported by process-aware information systems are subject to continuous and often subtle changes due to evolving operational,organizational,or regulatory factors.These changes,referred to as incremental con...Processes supported by process-aware information systems are subject to continuous and often subtle changes due to evolving operational,organizational,or regulatory factors.These changes,referred to as incremental concept drift,gradually alter the behavior or structure of processes,making their detection and localization a challenging task.Traditional process mining techniques frequently assume process stationarity and are limited in their ability to detect such drift,particularly from a control-flow perspective.The objective of this research is to develop an interpretable and robust framework capable of detecting and localizing incremental concept drift in event logs,with a specific emphasis on the structural evolution of control-flow semantics in processes.We propose DriftXMiner,a control-flow-aware hybrid framework that combines statistical,machine learning,and process model analysis techniques.The approach comprises three key components:(1)Cumulative Drift Scanner that tracks directional statistical deviations to detect early drift signals;(2)a Temporal Clustering and Drift-Aware Forest Ensemble(DAFE)to capture distributional and classification-level changes in process behavior;and(3)Petri net-based process model reconstruction,which enables the precise localization of structural drift using transition deviation metrics and replay fitness scores.Experimental validation on the BPI Challenge 2017 event log demonstrates that DriftXMiner effectively identifies and localizes gradual and incremental process drift over time.The framework achieves a detection accuracy of 92.5%,a localization precision of 90.3%,and an F1-score of 0.91,outperforming competitive baselines such as CUSUM+Histograms and ADWIN+Alpha Miner.Visual analyses further confirm that identified drift points align with transitions in control-flow models and behavioral cluster structures.DriftXMiner offers a novel and interpretable solution for incremental concept drift detection and localization in dynamic,process-aware systems.By integrating statistical signal accumulation,temporal behavior profiling,and structural process mining,the framework enables finegrained drift explanation and supports adaptive process intelligence in evolving environments.Its modular architecture supports extension to streaming data and real-time monitoring contexts.展开更多
One-dimensional ensemble dispersion entropy(EDE1D)is an effective nonlinear dynamic analysis method for complexity measurement of time series.However,it is only restricted to assessing the complexity of one-di-mension...One-dimensional ensemble dispersion entropy(EDE1D)is an effective nonlinear dynamic analysis method for complexity measurement of time series.However,it is only restricted to assessing the complexity of one-di-mensional time series(TS1d)with the extracted complexity features only at a single scale.Aiming at these problems,a new nonlinear dynamic analysis method termed two-dimensional composite multi-scale ensemble Gramian dispersion entropy(CMEGDE_(2D))is proposed in this paper.First,the TS_(1D) is transformed into a two-dimensional image(I_(2D))by using Gramian angular fields(GAF)with more internal data structures and geometri features,which preserve the global characteristics and time dependence of vibration signals.Second,the I2D is analyzed at multiple scales through the composite coarse-graining method,which overcomes the limitation of a single scale and provides greater stability compared to traditional coarse-graining methods.Subsequently,a new fault diagnosis method of rolling bearing is proposed based on the proposed CMEGDE_(2D) for fault feature ex-traction and the chicken swarm algorithm optimized support vector machine(CsO-SvM)for fault pattern identification.The simulation signals and two data sets of rolling bearings are utilized to verify the effectiveness of the proposed fault diagnosis method.The results demonstrate that the proposed method has stronger dis-crimination ability,higher fault diagnosis accuracy and better stability than the other compared methods.展开更多
The surge in smishing attacks underscores the urgent need for robust,real-time detection systems powered by advanced deep learning models.This paper introduces PhishNet,a novel ensemble learning framework that integra...The surge in smishing attacks underscores the urgent need for robust,real-time detection systems powered by advanced deep learning models.This paper introduces PhishNet,a novel ensemble learning framework that integrates transformer-based models(RoBERTa)and large language models(LLMs)(GPT-OSS 120B,LLaMA3.370B,and Qwen332B)to enhance smishing detection performance significantly.To mitigate class imbalance,we apply synthetic data augmentation using T5 and leverage various text preprocessing techniques.Our system employs a duallayer voting mechanism:weighted majority voting among LLMs and a final ensemble vote to classify messages as ham,spam,or smishing.Experimental results show an average accuracy improvement from 96%to 98.5%compared to the best standalone transformer,and from 93%to 98.5%when compared to LLMs across datasets.Furthermore,we present a real-time,user-friendly application to operationalize our detection model for practical use.PhishNet demonstrates superior scalability,usability,and detection accuracy,filling critical gaps in current smishing detection methodologies.展开更多
Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)t...Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)techniques for DDoS attack diagnosis normally apply network traffic statistical features such as packet sizes and inter-arrival times.However,such techniques sometimes fail to capture complicated relations among various traffic flows.In this paper,we present a new multi-scale ensemble strategy given the Graph Neural Networks(GNNs)for improving DDoS detection.Our technique divides traffic into macro-and micro-level elements,letting various GNN models to get the two corase-scale anomalies and subtle,stealthy attack models.Through modeling network traffic as graph-structured data,GNNs efficiently learn intricate relations among network entities.The proposed ensemble learning algorithm combines the results of several GNNs to improve generalization,robustness,and scalability.Extensive experiments on three benchmark datasets—UNSW-NB15,CICIDS2017,and CICDDoS2019—show that our approach outperforms traditional machine learning and deep learning models in detecting both high-rate and low-rate(stealthy)DDoS attacks,with significant improvements in accuracy and recall.These findings demonstrate the suggested method’s applicability and robustness for real-world implementation in contexts where several DDoS patterns coexist.展开更多
In recent years,ransomware attacks have become one of the most common and destructive types of cyberattacks.Their impact is significant on the operations,finances and reputation of affected companies.Despite the effor...In recent years,ransomware attacks have become one of the most common and destructive types of cyberattacks.Their impact is significant on the operations,finances and reputation of affected companies.Despite the efforts of researchers and security experts to protect information systems from these attacks,the threat persists and the proposed solutions are not able to significantly stop the spread of ransomware attacks.The latest remarkable achievements of large language models(LLMs)in NLP tasks have caught the attention of cybersecurity researchers to integrate thesemodels into security threat detection.Thesemodels offer high embedding capabilities,able to extract rich semantic representations and paving theway formore accurate and adaptive solutions.In this context,we propose a new approach for ransomware detection based on an ensemblemethod that leverages three distinctLLMembeddingmodels.This ensemble strategy takes advantage of the variety of embedding methods and the strengths of each model.In the proposed solution,each embedding model is associated with an independently trainedMLP classifier.The predictions obtained are then merged using a weighted voting technique,assigning each model an influence proportional to its performance.This approach makes it possible to exploit the complementarity of representations,improve detection accuracy and robustness,and offer a more reliable solution in the face of the growing diversity and complexity of modern ransomware.展开更多
Rice is one of the most important staple crops globally.Rice plant diseases can severely reduce crop yields and,in extreme cases,lead to total production loss.Early diagnosis enables timely intervention,mitigates dise...Rice is one of the most important staple crops globally.Rice plant diseases can severely reduce crop yields and,in extreme cases,lead to total production loss.Early diagnosis enables timely intervention,mitigates disease severity,supports effective treatment strategies,and reduces reliance on excessive pesticide use.Traditional machine learning approaches have been applied for automated rice disease diagnosis;however,these methods depend heavily on manual image preprocessing and handcrafted feature extraction,which are labor-intensive and time-consuming and often require domain expertise.Recently,end-to-end deep learning(DL) models have been introduced for this task,but they often lack robustness and generalizability across diverse datasets.To address these limitations,we propose a novel end-toend training framework for convolutional neural network(CNN) and attention-based model ensembles(E2ETCA).This framework integrates features from two state-of-the-art(SOTA) CNN models,Inception V3 and DenseNet-201,and an attention-based vision transformer(ViT) model.The fused features are passed through an additional fully connected layer with softmax activation for final classification.The entire process is trained end-to-end,enhancing its suitability for realworld deployment.Furthermore,we extract and analyze the learned features using a support vector machine(SVM),a traditional machine learning classifier,to provide comparative insights.We evaluate the proposed E2ETCA framework on three publicly available datasets,the Mendeley Rice Leaf Disease Image Samples dataset,the Kaggle Rice Diseases Image dataset,the Bangladesh Rice Research Institute dataset,and a combined version of all three.Using standard evaluation metrics(accuracy,precision,recall,and F1-score),our framework demonstrates superior performance compared to existing SOTA methods in rice disease diagnosis,with potential applicability to other agricultural disease detection tasks.展开更多
Optical non-reciprocity is a fundamental phenomenon in photonics.It is crucial for developing devices that rely on directional signal control,such as optical isolators and circulators.However,most research in this fie...Optical non-reciprocity is a fundamental phenomenon in photonics.It is crucial for developing devices that rely on directional signal control,such as optical isolators and circulators.However,most research in this field has focused on systems in equilibrium or steady states.In this work,we demonstrate a room-temperature Rydberg atomic platform where the unidirectional propagation of light acts as a switch to mediate time-crystalline-like collective oscillations through atomic synchronization.展开更多
Android smartphones have become an integral part of our daily lives,becoming targets for ransomware attacks.Such attacks encrypt user information and ask for payment to recover it.Conventional detection mechanisms,suc...Android smartphones have become an integral part of our daily lives,becoming targets for ransomware attacks.Such attacks encrypt user information and ask for payment to recover it.Conventional detection mechanisms,such as signature-based and heuristic techniques,often fail to detect new and polymorphic ransomware samples.To address this challenge,we employed various ensemble classifiers,such as Random Forest,Gradient Boosting,Bagging,and AutoML models.We aimed to showcase how AutoML can automate processes such as model selection,feature engineering,and hyperparameter optimization,to minimize manual effort while ensuring or enhancing performance compared to traditional approaches.We used this framework to test it with a publicly available dataset from the Kaggle repository,which contains features for Android ransomware network traffic.The dataset comprises 392,024 flow records,divided into eleven groups.There are ten classes for various ransomware types,including SVpeng,PornDroid,Koler,WannaLocker,and Lockerpin.There is also a class for regular traffic.We applied a three-step procedure to select themost relevant features:filter,wrapper,and embeddedmethods.The Bagging classifier was highly accurate,correctly getting 99.84%of the time.The FLAML AutoML framework was evenmore accurate,correctly getting 99.85%of the time.This is indicative of howwellAutoML performs in improving things with minimal human assistance.Our findings indicate that AutoML is an efficient,scalable,and flexible method to discover Android ransomware,and it will facilitate the development of next-generation intrusion detection systems.展开更多
In dry-coupled ultrasonic thickness measurement,thick rubber layers introduce high-amplitude parasitic echoes that obscure defect signals and degrade thickness accuracy.Existing methods struggle to resolve overlap-pin...In dry-coupled ultrasonic thickness measurement,thick rubber layers introduce high-amplitude parasitic echoes that obscure defect signals and degrade thickness accuracy.Existing methods struggle to resolve overlap-ping echoes under variable coupling conditions and non-stationary noise.This study proposes a novel dual-criterion framework integrating energy contribution and statistical impulsivity metrics to isolate specimen re-flections from coupling-layer interference.By decomposing A-scan signals into Intrinsic Mode Functions(IMFs),the framework employs energy contribution thresholds(>85%)and kurtosis indices(>3)to autonomously select IMFs containing valid specimen echoes.Hybrid time-frequency thresholding further suppresses interference through amplitude filtering and spectral focusing.Experimental results demonstrate the framework’s robustness,achieving 92.3%thickness accuracy for 5 mm steel specimens with 5 mm rubber coupling,outperforming conventional methods by up to 18.7%.The dual-criterion approach reduces operator dependency by 37%and maintainsΔT<0.03 mm under surface roughness up to 6.3μm,offering a practical solution for industrial nondestructive testing with thick dry-coupled interfaces.展开更多
This study was aimed to prepare landslide susceptibility maps for the Pithoragarh district in Uttarakhand,India,using advanced ensemble models that combined Radial Basis Function Networks(RBFN)with three ensemble lear...This study was aimed to prepare landslide susceptibility maps for the Pithoragarh district in Uttarakhand,India,using advanced ensemble models that combined Radial Basis Function Networks(RBFN)with three ensemble learning techniques:DAGGING(DG),MULTIBOOST(MB),and ADABOOST(AB).This combination resulted in three distinct ensemble models:DG-RBFN,MB-RBFN,and AB-RBFN.Additionally,a traditional weighted method,Information Value(IV),and a benchmark machine learning(ML)model,Multilayer Perceptron Neural Network(MLP),were employed for comparison and validation.The models were developed using ten landslide conditioning factors,which included slope,aspect,elevation,curvature,land cover,geomorphology,overburden depth,lithology,distance to rivers and distance to roads.These factors were instrumental in predicting the output variable,which was the probability of landslide occurrence.Statistical analysis of the models’performance indicated that the DG-RBFN model,with an Area Under ROC Curve(AUC)of 0.931,outperformed the other models.The AB-RBFN model achieved an AUC of 0.929,the MB-RBFN model had an AUC of 0.913,and the MLP model recorded an AUC of 0.926.These results suggest that the advanced ensemble ML model DG-RBFN was more accurate than traditional statistical model,single MLP model,and other ensemble models in preparing trustworthy landslide susceptibility maps,thereby enhancing land use planning and decision-making.展开更多
The burgeoning market for lithium-ion batteries has stimulated a growing need for more reliable battery performance monitoring. Accurate state-of-health(SOH) estimation is critical for ensuring battery operational per...The burgeoning market for lithium-ion batteries has stimulated a growing need for more reliable battery performance monitoring. Accurate state-of-health(SOH) estimation is critical for ensuring battery operational performance. Despite numerous data-driven methods reported in existing research for battery SOH estimation, these methods often exhibit inconsistent performance across different application scenarios. To address this issue and overcome the performance limitations of individual data-driven models,integrating multiple models for SOH estimation has received considerable attention. Ensemble learning(EL) typically leverages the strengths of multiple base models to achieve more robust and accurate outputs. However, the lack of a clear review of current research hinders the further development of ensemble methods in SOH estimation. Therefore, this paper comprehensively reviews multi-model ensemble learning methods for battery SOH estimation. First, existing ensemble methods are systematically categorized into 6 classes based on their combination strategies. Different realizations and underlying connections are meticulously analyzed for each category of EL methods, highlighting distinctions, innovations, and typical applications. Subsequently, these ensemble methods are comprehensively compared in terms of base models, combination strategies, and publication trends. Evaluations across 6 dimensions underscore the outstanding performance of stacking-based ensemble methods. Following this, these ensemble methods are further inspected from the perspectives of weighted ensemble and diversity, aiming to inspire potential approaches for enhancing ensemble performance. Moreover, addressing challenges such as base model selection, measuring model robustness and uncertainty, and interpretability of ensemble models in practical applications is emphasized. Finally, future research prospects are outlined, specifically noting that deep learning ensemble is poised to advance ensemble methods for battery SOH estimation. The convergence of advanced machine learning with ensemble learning is anticipated to yield valuable avenues for research. Accelerated research in ensemble learning holds promising prospects for achieving more accurate and reliable battery SOH estimation under real-world conditions.展开更多
文摘Diabetes is increasing commonly in people’s daily life and represents an extraordinary threat to human well-being.Machine Learning(ML)in the healthcare industry has recently made headlines.Several ML models are developed around different datasets for diabetic prediction.It is essential for ML models to predict diabetes accurately.Highly informative features of the dataset are vital to determine the capability factors of the model in the prediction of diabetes.Feature engineering(FE)is the way of taking forward in yielding highly informative features.Pima Indian Diabetes Dataset(PIDD)is used in this work,and the impact of informative features in ML models is experimented with and analyzed for the prediction of diabetes.Missing values(MV)and the effect of the imputation process in the data distribution of each feature are analyzed.Permutation importance and partial dependence are carried out extensively and the results revealed that Glucose(GLUC),Body Mass Index(BMI),and Insulin(INS)are highly informative features.Derived features are obtained for BMI and INS to add more information with its raw form.The ensemble classifier with an ensemble of AdaBoost(AB)and XGBoost(XB)is considered for the impact analysis of the proposed FE approach.The ensemble model performs well for the inclusion of derived features provided the high Diagnostics Odds Ratio(DOR)of 117.694.This shows a high margin of 8.2%when compared with the ensemble model with no derived features(DOR=96.306)included in the experiment.The inclusion of derived features with the FE approach of the current state-of-the-art made the ensemble model performs well with Sensitivity(0.793),Specificity(0.945),DOR(79.517),and False Omission Rate(0.090)which further improves the state-of-the-art results.
基金The authors would like to express their gratitude to Taif University,Taif,Saudi Arabia for providing administrative and technical support.This work was supported by the Taif University Researchers supporting Project number(TURSP-2020/254).
文摘This research work proposes a new stack-based generalization ensemble model to forecast the number of incidences of conjunctivitis disease.In addition to forecasting the occurrences of conjunctivitis incidences,the proposed model also improves performance by using the ensemble model.Weekly rate of acute Conjunctivitis per 1000 for Hong Kong is collected for the duration of the first week of January 2010 to the last week of December 2019.Pre-processing techniques such as imputation of missing values and logarithmic transformation are applied to pre-process the data sets.A stacked generalization ensemble model based on Auto-ARIMA(Autoregressive Integrated Moving Average),NNAR(Neural Network Autoregression),ETS(Exponential Smoothing),HW(Holt Winter)is proposed and applied on the dataset.Predictive analysis is conducted on the collected dataset of conjunctivitis disease,and further compared for different performance measures.The result shows that the RMSE(Root Mean Square Error),MAE(Mean Absolute Error),MAPE(Mean Absolute Percentage Error),ACF1(Auto Correlation Function)of the proposed ensemble is decreased significantly.Considering the RMSE,for instance,error values are reduced by 39.23%,9.13%,20.42%,and 17.13%in comparison to Auto-ARIMA,NAR,ETS,and HW model respectively.This research concludes that the accuracy of the forecasting of diseases can be significantly increased by applying the proposed stack generalization ensemble model as it minimizes the prediction error and hence provides better prediction trends as compared to Auto-ARIMA,NAR,ETS,and HW model applied discretely.
文摘Distribution transformers play a vital role in power distribution systems,and their reliable operation is crucial for grid stability.This study presents a simulation-based framework for active fault diagnosis and early warning of distribution transformers,integrating Sample Ensemble Learning(SEL)with a Self-Optimizing Support Vector Machine(SO-SVM).The SEL technique enhances data diversity and mitigates class imbalance,while SO-SVM adaptively tunes its hyperparameters to improve classification accuracy.A comprehensive transformer model was developed in MATLAB/Simulink to simulate diverse fault scenarios,including inter-turn winding faults,core saturation,and thermal aging.Feature vectors were extracted from voltage,current,and temperature measurements to train and validate the proposed hybrid model.Quantitative analysis shows that the SEL–SO-SVM framework achieves a classification accuracy of 97.8%,a precision of 96.5%,and an F1-score of 97.2%.Beyond classification,the model effectively identified incipient faults,providing an early warning lead time of up to 2.5 s before significant deviations in operational parameters.This predictive capability underscores its potential for preventing catastrophic transformer failures and enabling timely maintenance actions.The proposed approach demonstrates strong applicability for enhancing the reliability and operational safety of distribution transformers in simulated environments,offering a promising foundation for future real-time and field-level implementations.
基金the Deanship of Research and Graduate Studies at King Khalid University,KSA,for funding this work through the Large Research Project under grant number RGP2/164/46.
文摘Background:Stomach cancer(SC)is one of the most lethal malignancies worldwide due to late-stage diagnosis and limited treatment.The transcriptomic,epigenomic,and proteomic,etc.,omics datasets generated by high-throughput sequencing technology have become prominent in biomedical research,and they reveal molecular aspects of cancer diagnosis and therapy.Despite the development of advanced sequencing technology,the presence of high-dimensionality in multi-omics data makes it challenging to interpret the data.Methods:In this study,we introduce RankXLAN,an explainable ensemble-based multi-omics framework that integrates feature selection(FS),ensemble learning,bioinformatics,and in-silico validation for robust biomarker detection,potential therapeutic drug-repurposing candidates’identification,and classification of SC.To enhance the interpretability of the model,we incorporated explainable artificial intelligence(SHapley Additive exPlanations analysis),as well as accuracy,precision,F1-score,recall,cross-validation,specificity,likelihood ratio(LR)+,LR−,and Youden index results.Results:The experimental results showed that the top four FS algorithms achieved improved results when applied to the ensemble learning classification model.The proposed ensemble model produced an area under the curve(AUC)score of 0.994 for gene expression,0.97 for methylation,and 0.96 for miRNA expression data.Through the integration of bioinformatics and ML approach of the transcriptomic and epigenomic multi-omics dataset,we identified potential marker genes,namely,UBE2D2,HPCAL4,IGHA1,DPT,and FN3K.In-silico molecular docking revealed a strong binding affinity between ANKRD13C and the FDA-approved drug Everolimus(binding affinity−10.1 kcal/mol),identifying ANKRD13C as a potential therapeutic drug-repurposing target for SC.Conclusion:The proposed framework RankXLAN outperforms other existing frameworks for serum biomarker identification,therapeutic target identification,and SC classification with multi-omics datasets.
基金funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R104)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Modern intrusion detection systems(MIDS)face persistent challenges in coping with the rapid evolution of cyber threats,high-volume network traffic,and imbalanced datasets.Traditional models often lack the robustness and explainability required to detect novel and sophisticated attacks effectively.This study introduces an advanced,explainable machine learning framework for multi-class IDS using the KDD99 and IDS datasets,which reflects real-world network behavior through a blend of normal and diverse attack classes.The methodology begins with sophisticated data preprocessing,incorporating both RobustScaler and QuantileTransformer to address outliers and skewed feature distributions,ensuring standardized and model-ready inputs.Critical dimensionality reduction is achieved via the Harris Hawks Optimization(HHO)algorithm—a nature-inspired metaheuristic modeled on hawks’hunting strategies.HHO efficiently identifies the most informative features by optimizing a fitness function based on classification performance.Following feature selection,the SMOTE is applied to the training data to resolve class imbalance by synthetically augmenting underrepresented attack types.The stacked architecture is then employed,combining the strengths of XGBoost,SVM,and RF as base learners.This layered approach improves prediction robustness and generalization by balancing bias and variance across diverse classifiers.The model was evaluated using standard classification metrics:precision,recall,F1-score,and overall accuracy.The best overall performance was recorded with an accuracy of 99.44%for UNSW-NB15,demonstrating the model’s effectiveness.After balancing,the model demonstrated a clear improvement in detecting the attacks.We tested the model on four datasets to show the effectiveness of the proposed approach and performed the ablation study to check the effect of each parameter.Also,the proposed model is computationaly efficient.To support transparency and trust in decision-making,explainable AI(XAI)techniques are incorporated that provides both global and local insight into feature contributions,and offers intuitive visualizations for individual predictions.This makes it suitable for practical deployment in cybersecurity environments that demand both precision and accountability.
文摘The growing use of Portable Document Format(PDF)files across various sectors such as education,government,and business has inadvertently turned them into a major target for cyberattacks.Cybercriminals take advantage of the inherent flexibility and layered structure ofPDFs to inject malicious content,often employing advanced obfuscation techniques to evade detection by traditional signature-based security systems.These conventional methods are no longer adequate,especially against sophisticated threats like zero-day exploits and polymorphic malware.In response to these challenges,this study introduces a machine learning-based detection framework specifically designed to combat such threats.Central to the proposed solution is a stacked ensemble learning model that combines the strengths of four high-performing classifiers:Random Forest(RF),Extreme Gradient Boosting(XGB),LightGBM(LGBM),and CatBoost(CB).These models operate in parallel as base learners,each capturing different aspects of the data.Their outputs are then refined by a Gradient Boosting Classifier(GBC),which serves as a meta-learner to enhance prediction accuracy.To ensure the model remains both efficient and effective,Principal Component Analysis(PCA)is applied to reduce feature dimensionality while preserving critical information necessary for malware classification.The model is trained and validated using the CIC-Evasive PDFMalware2022 dataset,which includes a wide range of both malicious and benign PDF samples.The results demonstrate that the framework achieves impressive performance,with 97.10% accuracy and a 97.39% F1-score,surpassing several existing techniques.To enhance trust and interpretability,the system incorporates Local Interpretable Model-agnostic Explanations(LIME),which provides user-friendly insights into the rationale behind each prediction.This research emphasizes how the integration of ensemble learning,feature reduction,and explainable AI can lead to a practical and scalable solution for detecting complex PDF-based threats.The proposed framework lays the foundation for the next generation of intelligent,resilient cybersecurity systems that can address ever-evolving attack strategies.
文摘Background:Early detection of harmful brain activity in critically ill patients using electroencephalography(EEG)is vital for timely and effective clinical intervention.Automating EEG analysis with deep learning techniques holds significant promise for enhancing diagnostic efficiency and accuracy.Methods:We implemented EfficientNetB2,which leverages convolutional neural networks with a novel Temporal Squeeze-and-Excitation module to capture temporal EEG features,and WaveNet,a sequential model designed to effectively model temporal dependencies in EEG data using dilated causal convolutions and temporal self-attention.Both models were trained and evaluated using a publicly available EEG dataset,with performance assessed via 4-fold cross-validation and a step-wise learning rate reduction strategy.Results:Our results demonstrate a significant reduction in training loss from 0.6459 to 0.3055 and validation loss from 0.9602 to 0.5719 over six epochs.Consistent improvements were observed across cross-validation folds,highlighting the robustness of the models.Additionally,ensemble learning of the two architectures further enhanced classification performance.Conclusion:This comparative analysis sheds light on the strengths and limitations of EfficientNetB2 and WaveNet for automated harmful brain activity detection in EEG signals.The findings contribute to the advancement of reliable and efficient deep learning models,paving the way for their clinical application in managing critically ill patients.
基金Project supported by the National Natural Science Foundation of China(Grant No.62471180)。
文摘We propose a novel cooling protocol within a triple-Laguerre-Gaussian cavity optomechanical system,which is designed to suppress the thermal vibrations of a rotating mirror to reach its quantum ground state.The system incorporates two auxiliary cavities and an atomic ensemble coupled to a Laguerre-Gaussian rotational cavity.By carefully selecting system parameters,the cooling process of the rotating mirror is significantly enhanced,while the heating process is effectively suppressed,enabling efficient ground-state cooling even in the unresolved sideband regime.Compared to previous works,our scheme reduces the stringent restrictions on auxiliary systems,making it more experimentally feasible under broader parameter conditions.These findings provide a robust approach for achieving ground-state cooling in mechanical resonators.
基金supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.RS-2024-00406320)the Institute of Information&Communica-tions Technology Planning&Evaluation(IITP)-Innovative Human Resource Development for Local Intellectualization Program Grant funded by the Korea government(MSIT)(IITP-2026-RS-2023-00259678).
文摘Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional domain adaptation methods assume a single source domain,making them less suitable for modern deep learning settings that rely on diverse and large-scale datasets.To address this limitation,recent research has focused on Multi-Source Domain Adaptation(MSDA),which aims to learn effectively from multiple source domains.In this paper,we propose Efficient Domain Transition for Multi-source(EDTM),a novel and efficient framework designed to tackle two major challenges in existing MSDA approaches:(1)integrating knowledge across different source domains and(2)aligning label distributions between source and target domains.EDTM leverages an ensemble-based classifier expert mechanism to enhance the contribution of source domains that are more similar to the target domain.To further stabilize the learning process and improve performance,we incorporate imitation learning into the training of the target model.In addition,Maximum Classifier Discrepancy(MCD)is employed to align class-wise label distributions between the source and target domains.Experiments were conducted using Digits-Five,one of the most representative benchmark datasets for MSDA.The results show that EDTM consistently outperforms existing methods in terms of average classification accuracy.Notably,EDTM achieved significantly higher performance on target domains such as Modified National Institute of Standards and Technolog with blended background images(MNIST-M)and Street View House Numbers(SVHN)datasets,demonstrating enhanced generalization compared to baseline approaches.Furthermore,an ablation study analyzing the contribution of each loss component validated the effectiveness of the framework,highlighting the importance of each module in achieving optimal performance.
文摘Processes supported by process-aware information systems are subject to continuous and often subtle changes due to evolving operational,organizational,or regulatory factors.These changes,referred to as incremental concept drift,gradually alter the behavior or structure of processes,making their detection and localization a challenging task.Traditional process mining techniques frequently assume process stationarity and are limited in their ability to detect such drift,particularly from a control-flow perspective.The objective of this research is to develop an interpretable and robust framework capable of detecting and localizing incremental concept drift in event logs,with a specific emphasis on the structural evolution of control-flow semantics in processes.We propose DriftXMiner,a control-flow-aware hybrid framework that combines statistical,machine learning,and process model analysis techniques.The approach comprises three key components:(1)Cumulative Drift Scanner that tracks directional statistical deviations to detect early drift signals;(2)a Temporal Clustering and Drift-Aware Forest Ensemble(DAFE)to capture distributional and classification-level changes in process behavior;and(3)Petri net-based process model reconstruction,which enables the precise localization of structural drift using transition deviation metrics and replay fitness scores.Experimental validation on the BPI Challenge 2017 event log demonstrates that DriftXMiner effectively identifies and localizes gradual and incremental process drift over time.The framework achieves a detection accuracy of 92.5%,a localization precision of 90.3%,and an F1-score of 0.91,outperforming competitive baselines such as CUSUM+Histograms and ADWIN+Alpha Miner.Visual analyses further confirm that identified drift points align with transitions in control-flow models and behavioral cluster structures.DriftXMiner offers a novel and interpretable solution for incremental concept drift detection and localization in dynamic,process-aware systems.By integrating statistical signal accumulation,temporal behavior profiling,and structural process mining,the framework enables finegrained drift explanation and supports adaptive process intelligence in evolving environments.Its modular architecture supports extension to streaming data and real-time monitoring contexts.
基金Supported by the National Natural Science Foundation of China(Grant No.51975004)the Outstanding Youth Fund of Universities in Anhui Province of China(Grant No.2022AH020032).
文摘One-dimensional ensemble dispersion entropy(EDE1D)is an effective nonlinear dynamic analysis method for complexity measurement of time series.However,it is only restricted to assessing the complexity of one-di-mensional time series(TS1d)with the extracted complexity features only at a single scale.Aiming at these problems,a new nonlinear dynamic analysis method termed two-dimensional composite multi-scale ensemble Gramian dispersion entropy(CMEGDE_(2D))is proposed in this paper.First,the TS_(1D) is transformed into a two-dimensional image(I_(2D))by using Gramian angular fields(GAF)with more internal data structures and geometri features,which preserve the global characteristics and time dependence of vibration signals.Second,the I2D is analyzed at multiple scales through the composite coarse-graining method,which overcomes the limitation of a single scale and provides greater stability compared to traditional coarse-graining methods.Subsequently,a new fault diagnosis method of rolling bearing is proposed based on the proposed CMEGDE_(2D) for fault feature ex-traction and the chicken swarm algorithm optimized support vector machine(CsO-SvM)for fault pattern identification.The simulation signals and two data sets of rolling bearings are utilized to verify the effectiveness of the proposed fault diagnosis method.The results demonstrate that the proposed method has stronger dis-crimination ability,higher fault diagnosis accuracy and better stability than the other compared methods.
基金funded by the Deanship of Scientific Research(DSR)at King Abdulaziz University,Jeddah,under Grant No.(GPIP:1074-612-2024).
文摘The surge in smishing attacks underscores the urgent need for robust,real-time detection systems powered by advanced deep learning models.This paper introduces PhishNet,a novel ensemble learning framework that integrates transformer-based models(RoBERTa)and large language models(LLMs)(GPT-OSS 120B,LLaMA3.370B,and Qwen332B)to enhance smishing detection performance significantly.To mitigate class imbalance,we apply synthetic data augmentation using T5 and leverage various text preprocessing techniques.Our system employs a duallayer voting mechanism:weighted majority voting among LLMs and a final ensemble vote to classify messages as ham,spam,or smishing.Experimental results show an average accuracy improvement from 96%to 98.5%compared to the best standalone transformer,and from 93%to 98.5%when compared to LLMs across datasets.Furthermore,we present a real-time,user-friendly application to operationalize our detection model for practical use.PhishNet demonstrates superior scalability,usability,and detection accuracy,filling critical gaps in current smishing detection methodologies.
文摘Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)techniques for DDoS attack diagnosis normally apply network traffic statistical features such as packet sizes and inter-arrival times.However,such techniques sometimes fail to capture complicated relations among various traffic flows.In this paper,we present a new multi-scale ensemble strategy given the Graph Neural Networks(GNNs)for improving DDoS detection.Our technique divides traffic into macro-and micro-level elements,letting various GNN models to get the two corase-scale anomalies and subtle,stealthy attack models.Through modeling network traffic as graph-structured data,GNNs efficiently learn intricate relations among network entities.The proposed ensemble learning algorithm combines the results of several GNNs to improve generalization,robustness,and scalability.Extensive experiments on three benchmark datasets—UNSW-NB15,CICIDS2017,and CICDDoS2019—show that our approach outperforms traditional machine learning and deep learning models in detecting both high-rate and low-rate(stealthy)DDoS attacks,with significant improvements in accuracy and recall.These findings demonstrate the suggested method’s applicability and robustness for real-world implementation in contexts where several DDoS patterns coexist.
基金funded by the Deanship of Graduate Studies and Scientific Research at Jouf University under grant No.(DGSSR-2024-02-01176).
文摘In recent years,ransomware attacks have become one of the most common and destructive types of cyberattacks.Their impact is significant on the operations,finances and reputation of affected companies.Despite the efforts of researchers and security experts to protect information systems from these attacks,the threat persists and the proposed solutions are not able to significantly stop the spread of ransomware attacks.The latest remarkable achievements of large language models(LLMs)in NLP tasks have caught the attention of cybersecurity researchers to integrate thesemodels into security threat detection.Thesemodels offer high embedding capabilities,able to extract rich semantic representations and paving theway formore accurate and adaptive solutions.In this context,we propose a new approach for ransomware detection based on an ensemblemethod that leverages three distinctLLMembeddingmodels.This ensemble strategy takes advantage of the variety of embedding methods and the strengths of each model.In the proposed solution,each embedding model is associated with an independently trainedMLP classifier.The predictions obtained are then merged using a weighted voting technique,assigning each model an influence proportional to its performance.This approach makes it possible to exploit the complementarity of representations,improve detection accuracy and robustness,and offer a more reliable solution in the face of the growing diversity and complexity of modern ransomware.
基金the Begum Rokeya University,Rangpur,and the United Arab Emirates University,UAE for partially supporting this work。
文摘Rice is one of the most important staple crops globally.Rice plant diseases can severely reduce crop yields and,in extreme cases,lead to total production loss.Early diagnosis enables timely intervention,mitigates disease severity,supports effective treatment strategies,and reduces reliance on excessive pesticide use.Traditional machine learning approaches have been applied for automated rice disease diagnosis;however,these methods depend heavily on manual image preprocessing and handcrafted feature extraction,which are labor-intensive and time-consuming and often require domain expertise.Recently,end-to-end deep learning(DL) models have been introduced for this task,but they often lack robustness and generalizability across diverse datasets.To address these limitations,we propose a novel end-toend training framework for convolutional neural network(CNN) and attention-based model ensembles(E2ETCA).This framework integrates features from two state-of-the-art(SOTA) CNN models,Inception V3 and DenseNet-201,and an attention-based vision transformer(ViT) model.The fused features are passed through an additional fully connected layer with softmax activation for final classification.The entire process is trained end-to-end,enhancing its suitability for realworld deployment.Furthermore,we extract and analyze the learned features using a support vector machine(SVM),a traditional machine learning classifier,to provide comparative insights.We evaluate the proposed E2ETCA framework on three publicly available datasets,the Mendeley Rice Leaf Disease Image Samples dataset,the Kaggle Rice Diseases Image dataset,the Bangladesh Rice Research Institute dataset,and a combined version of all three.Using standard evaluation metrics(accuracy,precision,recall,and F1-score),our framework demonstrates superior performance compared to existing SOTA methods in rice disease diagnosis,with potential applicability to other agricultural disease detection tasks.
基金supported by the National Natural Science Foundation of China (Grant No.12274131)the Innovation Program for Quantum Science and Technology (Grant No.2024ZD0300101)。
文摘Optical non-reciprocity is a fundamental phenomenon in photonics.It is crucial for developing devices that rely on directional signal control,such as optical isolators and circulators.However,most research in this field has focused on systems in equilibrium or steady states.In this work,we demonstrate a room-temperature Rydberg atomic platform where the unidirectional propagation of light acts as a switch to mediate time-crystalline-like collective oscillations through atomic synchronization.
基金supported through theOngoing Research Funding Program(ORF-2025-498),King Saud University,Riyadh,Saudi Arabia.
文摘Android smartphones have become an integral part of our daily lives,becoming targets for ransomware attacks.Such attacks encrypt user information and ask for payment to recover it.Conventional detection mechanisms,such as signature-based and heuristic techniques,often fail to detect new and polymorphic ransomware samples.To address this challenge,we employed various ensemble classifiers,such as Random Forest,Gradient Boosting,Bagging,and AutoML models.We aimed to showcase how AutoML can automate processes such as model selection,feature engineering,and hyperparameter optimization,to minimize manual effort while ensuring or enhancing performance compared to traditional approaches.We used this framework to test it with a publicly available dataset from the Kaggle repository,which contains features for Android ransomware network traffic.The dataset comprises 392,024 flow records,divided into eleven groups.There are ten classes for various ransomware types,including SVpeng,PornDroid,Koler,WannaLocker,and Lockerpin.There is also a class for regular traffic.We applied a three-step procedure to select themost relevant features:filter,wrapper,and embeddedmethods.The Bagging classifier was highly accurate,correctly getting 99.84%of the time.The FLAML AutoML framework was evenmore accurate,correctly getting 99.85%of the time.This is indicative of howwellAutoML performs in improving things with minimal human assistance.Our findings indicate that AutoML is an efficient,scalable,and flexible method to discover Android ransomware,and it will facilitate the development of next-generation intrusion detection systems.
基金funded by the National Natural Science Foundation of China,grant number U24A20135Inner Mongolia Natural Science Foundation major project,grant number 2023ZD12+7 种基金Inner Mongolia Autonomous Region key research and development and achievement transformation plan project,grant number 2023YFHH0090Natural Science Foundation of Inner Mongolia,grant number 2022MS05006Inner Mongolia Autonomous Region Talent Development FundUniversity basic research business expenses,grant number 2023RCTD012University basic research business expenses,grant number 2023QNJS075Postgraduate Research Innovation Program and of Inner Mongolia Autonomous Region,grant number KC2024053BUniversity basic research business expenses,grant number 2024YXXS012National Key Laboratory of Special Vehicle Design and Manufacturing Integration Technology,grant number GZ2023KF012.
文摘In dry-coupled ultrasonic thickness measurement,thick rubber layers introduce high-amplitude parasitic echoes that obscure defect signals and degrade thickness accuracy.Existing methods struggle to resolve overlap-ping echoes under variable coupling conditions and non-stationary noise.This study proposes a novel dual-criterion framework integrating energy contribution and statistical impulsivity metrics to isolate specimen re-flections from coupling-layer interference.By decomposing A-scan signals into Intrinsic Mode Functions(IMFs),the framework employs energy contribution thresholds(>85%)and kurtosis indices(>3)to autonomously select IMFs containing valid specimen echoes.Hybrid time-frequency thresholding further suppresses interference through amplitude filtering and spectral focusing.Experimental results demonstrate the framework’s robustness,achieving 92.3%thickness accuracy for 5 mm steel specimens with 5 mm rubber coupling,outperforming conventional methods by up to 18.7%.The dual-criterion approach reduces operator dependency by 37%and maintainsΔT<0.03 mm under surface roughness up to 6.3μm,offering a practical solution for industrial nondestructive testing with thick dry-coupled interfaces.
基金the University of Transport Technology under the project entitled“Application of Machine Learning Algorithms in Landslide Susceptibility Mapping in Mountainous Areas”with grant number DTTD2022-16.
文摘This study was aimed to prepare landslide susceptibility maps for the Pithoragarh district in Uttarakhand,India,using advanced ensemble models that combined Radial Basis Function Networks(RBFN)with three ensemble learning techniques:DAGGING(DG),MULTIBOOST(MB),and ADABOOST(AB).This combination resulted in three distinct ensemble models:DG-RBFN,MB-RBFN,and AB-RBFN.Additionally,a traditional weighted method,Information Value(IV),and a benchmark machine learning(ML)model,Multilayer Perceptron Neural Network(MLP),were employed for comparison and validation.The models were developed using ten landslide conditioning factors,which included slope,aspect,elevation,curvature,land cover,geomorphology,overburden depth,lithology,distance to rivers and distance to roads.These factors were instrumental in predicting the output variable,which was the probability of landslide occurrence.Statistical analysis of the models’performance indicated that the DG-RBFN model,with an Area Under ROC Curve(AUC)of 0.931,outperformed the other models.The AB-RBFN model achieved an AUC of 0.929,the MB-RBFN model had an AUC of 0.913,and the MLP model recorded an AUC of 0.926.These results suggest that the advanced ensemble ML model DG-RBFN was more accurate than traditional statistical model,single MLP model,and other ensemble models in preparing trustworthy landslide susceptibility maps,thereby enhancing land use planning and decision-making.
基金National Natural Science Foundation of China (52075420)Fundamental Research Funds for the Central Universities (xzy022023049)National Key Research and Development Program of China (2023YFB3408600)。
文摘The burgeoning market for lithium-ion batteries has stimulated a growing need for more reliable battery performance monitoring. Accurate state-of-health(SOH) estimation is critical for ensuring battery operational performance. Despite numerous data-driven methods reported in existing research for battery SOH estimation, these methods often exhibit inconsistent performance across different application scenarios. To address this issue and overcome the performance limitations of individual data-driven models,integrating multiple models for SOH estimation has received considerable attention. Ensemble learning(EL) typically leverages the strengths of multiple base models to achieve more robust and accurate outputs. However, the lack of a clear review of current research hinders the further development of ensemble methods in SOH estimation. Therefore, this paper comprehensively reviews multi-model ensemble learning methods for battery SOH estimation. First, existing ensemble methods are systematically categorized into 6 classes based on their combination strategies. Different realizations and underlying connections are meticulously analyzed for each category of EL methods, highlighting distinctions, innovations, and typical applications. Subsequently, these ensemble methods are comprehensively compared in terms of base models, combination strategies, and publication trends. Evaluations across 6 dimensions underscore the outstanding performance of stacking-based ensemble methods. Following this, these ensemble methods are further inspected from the perspectives of weighted ensemble and diversity, aiming to inspire potential approaches for enhancing ensemble performance. Moreover, addressing challenges such as base model selection, measuring model robustness and uncertainty, and interpretability of ensemble models in practical applications is emphasized. Finally, future research prospects are outlined, specifically noting that deep learning ensemble is poised to advance ensemble methods for battery SOH estimation. The convergence of advanced machine learning with ensemble learning is anticipated to yield valuable avenues for research. Accelerated research in ensemble learning holds promising prospects for achieving more accurate and reliable battery SOH estimation under real-world conditions.
