Large-scale neural networks-based federated learning(FL)has gained public recognition for its effective capabilities in distributed training.Nonetheless,the open system architecture inherent to federated learning syst...Large-scale neural networks-based federated learning(FL)has gained public recognition for its effective capabilities in distributed training.Nonetheless,the open system architecture inherent to federated learning systems raises concerns regarding their vulnerability to potential attacks.Poisoning attacks turn into a major menace to federated learning on account of their concealed property and potent destructive force.By altering the local model during routine machine learning training,attackers can easily contaminate the global model.Traditional detection and aggregation solutions mitigate certain threats,but they are still insufficient to completely eliminate the influence generated by attackers.Therefore,federated unlearning that can remove unreliable models while maintaining the accuracy of the global model has become a solution.Unfortunately some existing federated unlearning approaches are rather difficult to be applied in large neural network models because of their high computational expenses.Hence,we propose SlideFU,an efficient anti-poisoning attack federated unlearning framework.The primary concept of SlideFU is to employ sliding window to construct the training process,where all operations are confined within the window.We design a malicious detection scheme based on principal component analysis(PCA),which calculates the trust factors between compressed models in a low-cost way to eliminate unreliable models.After confirming that the global model is under attack,the system activates the federated unlearning process,calibrates the gradients based on the updated direction of the calibration gradients.Experiments on two public datasets demonstrate that our scheme can recover a robust model with extremely high efficiency.展开更多
As a fundamental component in computer vision,edges can be categorized into four types based on discontinuities in reflectance,illumination,surface normal,or depth.While deep CNNs have significantly advanced generic e...As a fundamental component in computer vision,edges can be categorized into four types based on discontinuities in reflectance,illumination,surface normal,or depth.While deep CNNs have significantly advanced generic edge detection,real-time multi-class semantic edge detection under resource constraints remains challenging.To address this,we propose a lightweight framework based on PiDiNet that enables fine-grained semantic edge detection.Our model simultaneously predicts background and four edge categories from full-resolution inputs,balancing accuracy and efficiency.Key contributions include:a multi-channel output structure expanding binary edge prediction to five classes,supported by a deep supervision mechanism;a dynamic class-balancing strategy combining adaptive weighting with physical priors to handle extreme class imbalance;and maintained architectural efficiency enabling real-time inference.Extensive evaluations on BSDS-RIND show our approach achieves accuracy competitive with state-of-the-art methods while operating in real time.展开更多
Alzheimer’s Disease(AD)is a progressive neurodegenerative disorder that significantly affects cognitive function,making early and accurate diagnosis essential.Traditional Deep Learning(DL)-based approaches often stru...Alzheimer’s Disease(AD)is a progressive neurodegenerative disorder that significantly affects cognitive function,making early and accurate diagnosis essential.Traditional Deep Learning(DL)-based approaches often struggle with low-contrast MRI images,class imbalance,and suboptimal feature extraction.This paper develops a Hybrid DL system that unites MobileNetV2 with adaptive classification methods to boost Alzheimer’s diagnosis by processing MRI scans.Image enhancement is done using Contrast-Limited Adaptive Histogram Equalization(CLAHE)and Enhanced Super-Resolution Generative Adversarial Networks(ESRGAN).A classification robustness enhancement system integrates class weighting techniques and a Matthews Correlation Coefficient(MCC)-based evaluation method into the design.The trained and validated model gives a 98.88%accuracy rate and 0.9614 MCC score.We also performed a 10-fold cross-validation experiment with an average accuracy of 96.52%(±1.51),a loss of 0.1671,and an MCC score of 0.9429 across folds.The proposed framework outperforms the state-of-the-art models with a 98%weighted F1-score while decreasing misdiagnosis results for every AD stage.The model demonstrates apparent separation abilities between AD progression stages according to the results of the confusion matrix analysis.These results validate the effectiveness of hybrid DL models with adaptive preprocessing for early and reliable Alzheimer’s diagnosis,contributing to improved computer-aided diagnosis(CAD)systems in clinical practice.展开更多
The exponential growth of the Internet of Things(IoT)has introduced significant security challenges,with zero-day attacks emerging as one of the most critical and challenging threats.Traditional Machine Learning(ML)an...The exponential growth of the Internet of Things(IoT)has introduced significant security challenges,with zero-day attacks emerging as one of the most critical and challenging threats.Traditional Machine Learning(ML)and Deep Learning(DL)techniques have demonstrated promising early detection capabilities.However,their effectiveness is limited when handling the vast volumes of IoT-generated data due to scalability constraints,high computational costs,and the costly time-intensive process of data labeling.To address these challenges,this study proposes a Federated Learning(FL)framework that leverages collaborative and hybrid supervised learning to enhance cyber threat detection in IoT networks.By employing Deep Neural Networks(DNNs)and decentralized model training,the approach reduces computational complexity while improving detection accuracy.The proposed model demonstrates robust performance,achieving accuracies of 94.34%,99.95%,and 87.94%on the publicly available kitsune,Bot-IoT,and UNSW-NB15 datasets,respectively.Furthermore,its ability to detect zero-day attacks is validated through evaluations on two additional benchmark datasets,TON-IoT and IoT-23,using a Deep Federated Learning(DFL)framework,underscoring the generalization and effectiveness of the model in heterogeneous and decentralized IoT environments.Experimental results demonstrate superior performance over existing methods,establishing the proposed framework as an efficient and scalable solution for IoT security.展开更多
As an emerging discipline,machine learning has been widely used in artificial intelligence,education,meteorology and other fields.In the training of machine learning models,trainers need to use a large amount of pract...As an emerging discipline,machine learning has been widely used in artificial intelligence,education,meteorology and other fields.In the training of machine learning models,trainers need to use a large amount of practical data,which inevitably involves user privacy.Besides,by polluting the training data,a malicious adversary can poison the model,thus compromising model security.The data provider hopes that the model trainer can prove to them the confidentiality of the model.Trainer will be required to withdraw data when the trust collapses.In the meantime,trainers hope to forget the injected data to regain security when finding crafted poisoned data after the model training.Therefore,we focus on forgetting systems,the process of which we call machine unlearning,capable of forgetting specific data entirely and efficiently.In this paper,we present the first comprehensive survey of this realm.We summarize and categorize existing machine unlearning methods based on their characteristics and analyze the relation between machine unlearning and relevant fields(e.g.,inference attacks and data poisoning attacks).Finally,we briefly conclude the existing research directions.展开更多
The aim of the article is to explore the relation among capitalism,creative economy,and the end of rest in Gustavo Vinagre’s movie Unlearning to Sleep.The main argument indicates that,in the context of the imperative...The aim of the article is to explore the relation among capitalism,creative economy,and the end of rest in Gustavo Vinagre’s movie Unlearning to Sleep.The main argument indicates that,in the context of the imperatives within the inhumane temporalities of the 24/7 society,sleep and rest may represent an inevitable and anomalous resistance to the demands of the capitalist order in which creative economy is immersed and exposed in the movie.展开更多
Cyberbullying on social media poses significant psychological risks,yet most detection systems over-simplify the task by focusing on binary classification,ignoring nuanced categories like passive-aggressive remarks or...Cyberbullying on social media poses significant psychological risks,yet most detection systems over-simplify the task by focusing on binary classification,ignoring nuanced categories like passive-aggressive remarks or indirect slurs.To address this gap,we propose a hybrid framework combining Term Frequency-Inverse Document Frequency(TF-IDF),word-to-vector(Word2Vec),and Bidirectional Encoder Representations from Transformers(BERT)based models for multi-class cyberbullying detection.Our approach integrates TF-IDF for lexical specificity and Word2Vec for semantic relationships,fused with BERT’s contextual embeddings to capture syntactic and semantic complexities.We evaluate the framework on a publicly available dataset of 47,000 annotated social media posts across five cyberbullying categories:age,ethnicity,gender,religion,and indirect aggression.Among BERT variants tested,BERT Base Un-Cased achieved the highest performance with 93%accuracy(standard deviation across±1%5-fold cross-validation)and an average AUC of 0.96,outperforming standalone TF-IDF(78%)and Word2Vec(82%)models.Notably,it achieved near-perfect AUC scores(0.99)for age and ethnicity-based bullying.A comparative analysis with state-of-the-art benchmarks,including Generative Pre-trained Transformer 2(GPT-2)and Text-to-Text Transfer Transformer(T5)models highlights BERT’s superiority in handling ambiguous language.This work advances cyberbullying detection by demonstrating how hybrid feature extraction and transformer models improve multi-class classification,offering a scalable solution for moderating nuanced harmful content.展开更多
Human activity recognition is a significant area of research in artificial intelligence for surveillance,healthcare,sports,and human-computer interaction applications.The article benchmarks the performance of You Only...Human activity recognition is a significant area of research in artificial intelligence for surveillance,healthcare,sports,and human-computer interaction applications.The article benchmarks the performance of You Only Look Once version 11-based(YOLOv11-based)architecture for multi-class human activity recognition.The article benchmarks the performance of You Only Look Once version 11-based(YOLOv11-based)architecture for multi-class human activity recognition.The dataset consists of 14,186 images across 19 activity classes,from dynamic activities such as running and swimming to static activities such as sitting and sleeping.Preprocessing included resizing all images to 512512 pixels,annotating them in YOLO’s bounding box format,and applying data augmentation methods such as flipping,rotation,and cropping to enhance model generalization.The proposed model was trained for 100 epochs with adaptive learning rate methods and hyperparameter optimization for performance improvement,with a mAP@0.5 of 74.93%and a mAP@0.5-0.95 of 64.11%,outperforming previous versions of YOLO(v10,v9,and v8)and general-purpose architectures like ResNet50 and EfficientNet.It exhibited improved precision and recall for all activity classes with high precision values of 0.76 for running,0.79 for swimming,0.80 for sitting,and 0.81 for sleeping,and was tested for real-time deployment with an inference time of 8.9 ms per image,being computationally light.Proposed YOLOv11’s improvements are attributed to architectural advancements like a more complex feature extraction process,better attention modules,and an anchor-free detection mechanism.While YOLOv10 was extremely stable in static activity recognition,YOLOv9 performed well in dynamic environments but suffered from overfitting,and YOLOv8,while being a decent baseline,failed to differentiate between overlapping static activities.The experimental results determine proposed YOLOv11 to be the most appropriate model,providing an ideal balance between accuracy,computational efficiency,and robustness for real-world deployment.Nevertheless,there exist certain issues to be addressed,particularly in discriminating against visually similar activities and the use of publicly available datasets.Future research will entail the inclusion of 3D data and multimodal sensor inputs,such as depth and motion information,for enhancing recognition accuracy and generalizability to challenging real-world environments.展开更多
This paper considers the problem of time varying congestion pricing to determine optimal time-varying tolls at peak periods for a queuing network with the interactions between buses and private cars.Through the combin...This paper considers the problem of time varying congestion pricing to determine optimal time-varying tolls at peak periods for a queuing network with the interactions between buses and private cars.Through the combined applications of the space-time expanded network（STEN） and the conventional network equilibrium modeling techniques,a multi-class,multi-mode and multi-criteria traffic network equilibrium model is developed.Travelers of different classes have distinctive value of times（VOTs）,and travelers from the same class perceive their travel disutility or generalized costs on a route according to different weights of travel time and travel costs.Moreover,the symmetric cost function model is extended to deal with the interactions between buses and private cars.It is found that there exists a uniform（anonymous） link toll pattern which can drive a multi-class,multi-mode and multi-criteria user equilibrium flow pattern to a system optimum when the system＇s objective function is measured in terms of money.It is also found that the marginal cost pricing models with a symmetric travel cost function do not reflect the interactions between traffic flows of different road sections,and the obtained congestion pricing toll is smaller than the real value.展开更多
Machine Unlearning(MU)has emerged as a promising technique for aligning large language models(LLMs)with safety requirements to steer them forgetting specific harmful contents.Despite the significant progress in previo...Machine Unlearning(MU)has emerged as a promising technique for aligning large language models(LLMs)with safety requirements to steer them forgetting specific harmful contents.Despite the significant progress in previous studies,we argue that the current evaluation criteria,which solely focus on safety evaluation,are actually impractical and biased,leading to concerns about the true effectiveness of MU techniques.To address this,we propose to comprehensively evaluate LLMs after MU from three aspects:safety,over-safety,and general utility.Specifically,a novel benchmark MUBENCH with 18 related datasets is first constructed,where the safety is measured with both vanilla harmful inputs and 10 types of jailbreak attacks.Furthermore,we examine whether MU introduces side effects,focusing on over-safety and utility-loss.Extensive experiments are performed on 3 popular LLMs with 7 recent MU methods.The results highlight a challenging trilemma in safety alignment without side effects,indicating that there is still considerable room for further exploration.MUBENCH serves as a comprehensive benchmark,fostering future research on MU for safety alignment of LLMs.展开更多
Aiming at the problem of imbalance between detection accuracy and algorithm model lightweight in UAV aerial image target detection algorithm,a lightweight multi-category abnormal behavior detection algorithm based on ...Aiming at the problem of imbalance between detection accuracy and algorithm model lightweight in UAV aerial image target detection algorithm,a lightweight multi-category abnormal behavior detection algorithm based on improved YOLOv11n is designed.By integrating multi-head grouped self-attention mechanism and Partial-Conv,a two-way feature grouping fusion module(DFPF)was designed,which carried out effective channel segmentation and fusion strategies to reduce redundant calculations andmemory access.C3K2 module was improved,and then unstructured pruning and feature distillation technologywere used.The algorithmmodel is lightweight,and the feature extraction ability for airborne visual abnormal behavior targets is strengthened,and the computational efficiency of the model is improved.Finally,we test the generalization of the baseline model and the improved model on the VisDrone2019 dataset.The results show that com-pared with the baseline model,the detection accuracy of the final improved model on the airborne visual abnormal behavior dataset is improved from 90.2% to 94.8%,and the model parameters are reduced by 50.9% to meet the detection requirements of high efficiency and high precision.The detection accuracy of the improved model on the Vis-Drone2019 public dataset is 1.3% higher than that of the baseline model,indicating the effectiveness of the improved method in this paper.展开更多
Aiming at the limitations of rapid fault diagnosis of blast furnace, a novel strategy based on cost-conscious least squares support vector machine （LS-SVM） is proposed to solve this problem. Firstly, modified discre...Aiming at the limitations of rapid fault diagnosis of blast furnace, a novel strategy based on cost-conscious least squares support vector machine （LS-SVM） is proposed to solve this problem. Firstly, modified discrete particle swarm optimization is applied to optimize the feature selection and the LS-SVM parameters. Secondly, cost-con- scious formula is presented for fitness function and it contains in detail training time, recognition accuracy and the feature selection. The CLS-SVM algorithm is presented to increase the performance of the LS-SVM classifier. The new method can select the best fault features in much shorter time and have fewer support vectbrs and better general- ization performance in the application of fault diagnosis of the blast furnace. Thirdly, a gradual change binary tree is established for blast furnace faults diagnosis. It is a multi-class classification method based on center-of-gravity formula distance of cluster. A gradual change classification percentage ia used to select sample randomly. The proposed new metbod raises the sped of diagnosis, optimizes the classifieation scraraey and has good generalization ability for fault diagnosis of the application of blast furnace.展开更多
We present a high-resolution relaxation scheme for a multi-class Lighthill-Whitham-Richards (MCLWR) traffic flow model. This scheme is based on high-order reconstruction for spatial discretization and an implicit-expl...We present a high-resolution relaxation scheme for a multi-class Lighthill-Whitham-Richards (MCLWR) traffic flow model. This scheme is based on high-order reconstruction for spatial discretization and an implicit-explicit Runge-Kutta method for time integration. The resulting method retains the simplicity of the relaxation schemes. There is no need to involve Riemann solvers and characteristic decomposition. Even the computation of the eigenvalues is not required. This makes the scheme particularly well suited for the MCLWR model in which the analytical expressions of the eigenvalues are difficult to obtain for more than four classes of road users. The numerical results illustrate the effectiveness of the presented method.展开更多
Hierarchical Support Vector Machine （H-SVM） is faster in training and classification than other usual multi-class SVMs such as ＂1-V-R＂and ＂1-V-1＂. In this paper, a new multi-class fault diagnosis algorithm based...Hierarchical Support Vector Machine （H-SVM） is faster in training and classification than other usual multi-class SVMs such as ＂1-V-R＂and ＂1-V-1＂. In this paper, a new multi-class fault diagnosis algorithm based on H-SVM is proposed and applied to aero-engine. Before SVM training, the training data are first clustered according to their class-center Euclid distances in some feature spaces. The samples which have close distances are divided into the same sub-classes for training, and this makes the H-SVM have reasonable hierarchical construction and good generalization performance. Instead of the common C-SVM, the v-SVM is selected as the binary classifier, in which the parameter v varies only from 0 to 1 and can be determined more easily. The simulation results show that the designed H-SVMs can fast diagnose the multi-class single faults and combination faults for the gas path components of an aero-engine. The fault classifiers have good diagnosis accuracy and can keep robust even when the measurement inputs are disturbed by noises.展开更多
Multi-source multi-class classification methods based on multi-class Support Vector Machines and data fusion strategies are proposed in this paper. The centralized and distributed fusion schemes are applied to combine...Multi-source multi-class classification methods based on multi-class Support Vector Machines and data fusion strategies are proposed in this paper. The centralized and distributed fusion schemes are applied to combine information from several data sources. In the centralized scheme, all information from several data sources is centralized to construct an input space. Then a multi-class Support Vector Machine classifier is trained. In the distributed schemes, the individual data sources are proc-essed separately and modelled by using the multi-class Support Vector Machine. Then new data fusion strategies are proposed to combine the information from the individual multi-class Support Vector Machine models. Our proposed fusion strategies take into account that an Support Vector Machine (SVM) classifier achieves classification by finding the optimal classification hyperplane with maximal margin. The proposed methods are applied for fault diagnosis of a diesel engine. The experimental results showed that almost all the proposed approaches can largely improve the diagnostic accuracy. The robustness of diagnosis is also improved because of the implementation of data fusion strategies. The proposed methods can also be applied in other fields.展开更多
基金supported in part by the National Social Science Foundation of China under Grant 20BTQ058in part by the Natural Science Foundation of Hunan Province under Grant 2023JJ50033.
文摘Large-scale neural networks-based federated learning(FL)has gained public recognition for its effective capabilities in distributed training.Nonetheless,the open system architecture inherent to federated learning systems raises concerns regarding their vulnerability to potential attacks.Poisoning attacks turn into a major menace to federated learning on account of their concealed property and potent destructive force.By altering the local model during routine machine learning training,attackers can easily contaminate the global model.Traditional detection and aggregation solutions mitigate certain threats,but they are still insufficient to completely eliminate the influence generated by attackers.Therefore,federated unlearning that can remove unreliable models while maintaining the accuracy of the global model has become a solution.Unfortunately some existing federated unlearning approaches are rather difficult to be applied in large neural network models because of their high computational expenses.Hence,we propose SlideFU,an efficient anti-poisoning attack federated unlearning framework.The primary concept of SlideFU is to employ sliding window to construct the training process,where all operations are confined within the window.We design a malicious detection scheme based on principal component analysis(PCA),which calculates the trust factors between compressed models in a low-cost way to eliminate unreliable models.After confirming that the global model is under attack,the system activates the federated unlearning process,calibrates the gradients based on the updated direction of the calibration gradients.Experiments on two public datasets demonstrate that our scheme can recover a robust model with extremely high efficiency.
基金supported by the National Natural Science Foundation of China 62402171.
文摘As a fundamental component in computer vision,edges can be categorized into four types based on discontinuities in reflectance,illumination,surface normal,or depth.While deep CNNs have significantly advanced generic edge detection,real-time multi-class semantic edge detection under resource constraints remains challenging.To address this,we propose a lightweight framework based on PiDiNet that enables fine-grained semantic edge detection.Our model simultaneously predicts background and four edge categories from full-resolution inputs,balancing accuracy and efficiency.Key contributions include:a multi-channel output structure expanding binary edge prediction to five classes,supported by a deep supervision mechanism;a dynamic class-balancing strategy combining adaptive weighting with physical priors to handle extreme class imbalance;and maintained architectural efficiency enabling real-time inference.Extensive evaluations on BSDS-RIND show our approach achieves accuracy competitive with state-of-the-art methods while operating in real time.
基金funded by the Deanship of Graduate Studies and Scientific Research at Jouf University under grant No.(DGSSR-2025-02-01295).
文摘Alzheimer’s Disease(AD)is a progressive neurodegenerative disorder that significantly affects cognitive function,making early and accurate diagnosis essential.Traditional Deep Learning(DL)-based approaches often struggle with low-contrast MRI images,class imbalance,and suboptimal feature extraction.This paper develops a Hybrid DL system that unites MobileNetV2 with adaptive classification methods to boost Alzheimer’s diagnosis by processing MRI scans.Image enhancement is done using Contrast-Limited Adaptive Histogram Equalization(CLAHE)and Enhanced Super-Resolution Generative Adversarial Networks(ESRGAN).A classification robustness enhancement system integrates class weighting techniques and a Matthews Correlation Coefficient(MCC)-based evaluation method into the design.The trained and validated model gives a 98.88%accuracy rate and 0.9614 MCC score.We also performed a 10-fold cross-validation experiment with an average accuracy of 96.52%(±1.51),a loss of 0.1671,and an MCC score of 0.9429 across folds.The proposed framework outperforms the state-of-the-art models with a 98%weighted F1-score while decreasing misdiagnosis results for every AD stage.The model demonstrates apparent separation abilities between AD progression stages according to the results of the confusion matrix analysis.These results validate the effectiveness of hybrid DL models with adaptive preprocessing for early and reliable Alzheimer’s diagnosis,contributing to improved computer-aided diagnosis(CAD)systems in clinical practice.
基金supported by Princess Nourah bint Abdulrahman University Researchers Supporting Project Number(PNURSP2025R97)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘The exponential growth of the Internet of Things(IoT)has introduced significant security challenges,with zero-day attacks emerging as one of the most critical and challenging threats.Traditional Machine Learning(ML)and Deep Learning(DL)techniques have demonstrated promising early detection capabilities.However,their effectiveness is limited when handling the vast volumes of IoT-generated data due to scalability constraints,high computational costs,and the costly time-intensive process of data labeling.To address these challenges,this study proposes a Federated Learning(FL)framework that leverages collaborative and hybrid supervised learning to enhance cyber threat detection in IoT networks.By employing Deep Neural Networks(DNNs)and decentralized model training,the approach reduces computational complexity while improving detection accuracy.The proposed model demonstrates robust performance,achieving accuracies of 94.34%,99.95%,and 87.94%on the publicly available kitsune,Bot-IoT,and UNSW-NB15 datasets,respectively.Furthermore,its ability to detect zero-day attacks is validated through evaluations on two additional benchmark datasets,TON-IoT and IoT-23,using a Deep Federated Learning(DFL)framework,underscoring the generalization and effectiveness of the model in heterogeneous and decentralized IoT environments.Experimental results demonstrate superior performance over existing methods,establishing the proposed framework as an efficient and scalable solution for IoT security.
基金supported by the National Key Research and Development Program of China(2020YFC2003404)the National Natura Science Foundation of China(No.62072465,62172155,62102425,62102429)+1 种基金the Science and Technology Innovation Program of Hunan Province(Nos.2022RC3061,2021RC2071)the Natural Science Foundation of Hunan Province(No.2022JJ40564).
文摘As an emerging discipline,machine learning has been widely used in artificial intelligence,education,meteorology and other fields.In the training of machine learning models,trainers need to use a large amount of practical data,which inevitably involves user privacy.Besides,by polluting the training data,a malicious adversary can poison the model,thus compromising model security.The data provider hopes that the model trainer can prove to them the confidentiality of the model.Trainer will be required to withdraw data when the trust collapses.In the meantime,trainers hope to forget the injected data to regain security when finding crafted poisoned data after the model training.Therefore,we focus on forgetting systems,the process of which we call machine unlearning,capable of forgetting specific data entirely and efficiently.In this paper,we present the first comprehensive survey of this realm.We summarize and categorize existing machine unlearning methods based on their characteristics and analyze the relation between machine unlearning and relevant fields(e.g.,inference attacks and data poisoning attacks).Finally,we briefly conclude the existing research directions.
文摘The aim of the article is to explore the relation among capitalism,creative economy,and the end of rest in Gustavo Vinagre’s movie Unlearning to Sleep.The main argument indicates that,in the context of the imperatives within the inhumane temporalities of the 24/7 society,sleep and rest may represent an inevitable and anomalous resistance to the demands of the capitalist order in which creative economy is immersed and exposed in the movie.
基金funded by Scientific Research Deanship at University of Hail-Saudi Arabia through Project Number RG-23092.
文摘Cyberbullying on social media poses significant psychological risks,yet most detection systems over-simplify the task by focusing on binary classification,ignoring nuanced categories like passive-aggressive remarks or indirect slurs.To address this gap,we propose a hybrid framework combining Term Frequency-Inverse Document Frequency(TF-IDF),word-to-vector(Word2Vec),and Bidirectional Encoder Representations from Transformers(BERT)based models for multi-class cyberbullying detection.Our approach integrates TF-IDF for lexical specificity and Word2Vec for semantic relationships,fused with BERT’s contextual embeddings to capture syntactic and semantic complexities.We evaluate the framework on a publicly available dataset of 47,000 annotated social media posts across five cyberbullying categories:age,ethnicity,gender,religion,and indirect aggression.Among BERT variants tested,BERT Base Un-Cased achieved the highest performance with 93%accuracy(standard deviation across±1%5-fold cross-validation)and an average AUC of 0.96,outperforming standalone TF-IDF(78%)and Word2Vec(82%)models.Notably,it achieved near-perfect AUC scores(0.99)for age and ethnicity-based bullying.A comparative analysis with state-of-the-art benchmarks,including Generative Pre-trained Transformer 2(GPT-2)and Text-to-Text Transfer Transformer(T5)models highlights BERT’s superiority in handling ambiguous language.This work advances cyberbullying detection by demonstrating how hybrid feature extraction and transformer models improve multi-class classification,offering a scalable solution for moderating nuanced harmful content.
基金supported by King Saud University,Riyadh,Saudi Arabia,under Ongoing Research Funding Program(ORF-2025-951).
文摘Human activity recognition is a significant area of research in artificial intelligence for surveillance,healthcare,sports,and human-computer interaction applications.The article benchmarks the performance of You Only Look Once version 11-based(YOLOv11-based)architecture for multi-class human activity recognition.The article benchmarks the performance of You Only Look Once version 11-based(YOLOv11-based)architecture for multi-class human activity recognition.The dataset consists of 14,186 images across 19 activity classes,from dynamic activities such as running and swimming to static activities such as sitting and sleeping.Preprocessing included resizing all images to 512512 pixels,annotating them in YOLO’s bounding box format,and applying data augmentation methods such as flipping,rotation,and cropping to enhance model generalization.The proposed model was trained for 100 epochs with adaptive learning rate methods and hyperparameter optimization for performance improvement,with a mAP@0.5 of 74.93%and a mAP@0.5-0.95 of 64.11%,outperforming previous versions of YOLO(v10,v9,and v8)and general-purpose architectures like ResNet50 and EfficientNet.It exhibited improved precision and recall for all activity classes with high precision values of 0.76 for running,0.79 for swimming,0.80 for sitting,and 0.81 for sleeping,and was tested for real-time deployment with an inference time of 8.9 ms per image,being computationally light.Proposed YOLOv11’s improvements are attributed to architectural advancements like a more complex feature extraction process,better attention modules,and an anchor-free detection mechanism.While YOLOv10 was extremely stable in static activity recognition,YOLOv9 performed well in dynamic environments but suffered from overfitting,and YOLOv8,while being a decent baseline,failed to differentiate between overlapping static activities.The experimental results determine proposed YOLOv11 to be the most appropriate model,providing an ideal balance between accuracy,computational efficiency,and robustness for real-world deployment.Nevertheless,there exist certain issues to be addressed,particularly in discriminating against visually similar activities and the use of publicly available datasets.Future research will entail the inclusion of 3D data and multimodal sensor inputs,such as depth and motion information,for enhancing recognition accuracy and generalizability to challenging real-world environments.
基金The National High Technology Research and Development Program of China (863 Program) (No. 2007AA11Z202)the National Key Technology R & D Program of China during the 11th Five-Year Plan Period(No. 2006BAJ18B03)the Fundamental Research Funds for the Central Universities (No. DUT10RC(3) 112)
文摘This paper considers the problem of time varying congestion pricing to determine optimal time-varying tolls at peak periods for a queuing network with the interactions between buses and private cars.Through the combined applications of the space-time expanded network（STEN） and the conventional network equilibrium modeling techniques,a multi-class,multi-mode and multi-criteria traffic network equilibrium model is developed.Travelers of different classes have distinctive value of times（VOTs）,and travelers from the same class perceive their travel disutility or generalized costs on a route according to different weights of travel time and travel costs.Moreover,the symmetric cost function model is extended to deal with the interactions between buses and private cars.It is found that there exists a uniform（anonymous） link toll pattern which can drive a multi-class,multi-mode and multi-criteria user equilibrium flow pattern to a system optimum when the system＇s objective function is measured in terms of money.It is also found that the marginal cost pricing models with a symmetric travel cost function do not reflect the interactions between traffic flows of different road sections,and the obtained congestion pricing toll is smaller than the real value.
基金supported by the National Natural Science Foundation of China(Grant No.62176078)the Fundamental Research Funds for the Central Universities(2022FRFK060002).
文摘Machine Unlearning(MU)has emerged as a promising technique for aligning large language models(LLMs)with safety requirements to steer them forgetting specific harmful contents.Despite the significant progress in previous studies,we argue that the current evaluation criteria,which solely focus on safety evaluation,are actually impractical and biased,leading to concerns about the true effectiveness of MU techniques.To address this,we propose to comprehensively evaluate LLMs after MU from three aspects:safety,over-safety,and general utility.Specifically,a novel benchmark MUBENCH with 18 related datasets is first constructed,where the safety is measured with both vanilla harmful inputs and 10 types of jailbreak attacks.Furthermore,we examine whether MU introduces side effects,focusing on over-safety and utility-loss.Extensive experiments are performed on 3 popular LLMs with 7 recent MU methods.The results highlight a challenging trilemma in safety alignment without side effects,indicating that there is still considerable room for further exploration.MUBENCH serves as a comprehensive benchmark,fostering future research on MU for safety alignment of LLMs.
基金supported by y the Applied Research Advancement Project in Engineering University of PAP(WYY202304)Research and Innovation Team Project in Engineering University of PAP(KYTD202306)Funding for postgraduate education and teaching.
文摘Aiming at the problem of imbalance between detection accuracy and algorithm model lightweight in UAV aerial image target detection algorithm,a lightweight multi-category abnormal behavior detection algorithm based on improved YOLOv11n is designed.By integrating multi-head grouped self-attention mechanism and Partial-Conv,a two-way feature grouping fusion module(DFPF)was designed,which carried out effective channel segmentation and fusion strategies to reduce redundant calculations andmemory access.C3K2 module was improved,and then unstructured pruning and feature distillation technologywere used.The algorithmmodel is lightweight,and the feature extraction ability for airborne visual abnormal behavior targets is strengthened,and the computational efficiency of the model is improved.Finally,we test the generalization of the baseline model and the improved model on the VisDrone2019 dataset.The results show that com-pared with the baseline model,the detection accuracy of the final improved model on the airborne visual abnormal behavior dataset is improved from 90.2% to 94.8%,and the model parameters are reduced by 50.9% to meet the detection requirements of high efficiency and high precision.The detection accuracy of the improved model on the Vis-Drone2019 public dataset is 1.3% higher than that of the baseline model,indicating the effectiveness of the improved method in this paper.
基金Item Sponsored by National Natural Science Foundation of China(60843007,61050006)
文摘Aiming at the limitations of rapid fault diagnosis of blast furnace, a novel strategy based on cost-conscious least squares support vector machine （LS-SVM） is proposed to solve this problem. Firstly, modified discrete particle swarm optimization is applied to optimize the feature selection and the LS-SVM parameters. Secondly, cost-con- scious formula is presented for fitness function and it contains in detail training time, recognition accuracy and the feature selection. The CLS-SVM algorithm is presented to increase the performance of the LS-SVM classifier. The new method can select the best fault features in much shorter time and have fewer support vectbrs and better general- ization performance in the application of fault diagnosis of the blast furnace. Thirdly, a gradual change binary tree is established for blast furnace faults diagnosis. It is a multi-class classification method based on center-of-gravity formula distance of cluster. A gradual change classification percentage ia used to select sample randomly. The proposed new metbod raises the sped of diagnosis, optimizes the classifieation scraraey and has good generalization ability for fault diagnosis of the application of blast furnace.
基金Project supported by the Aoxiang Project and the Scientific and Technological Innovation Foundation of Northwestern Polytechnical University, China (No 2007KJ01011)
文摘We present a high-resolution relaxation scheme for a multi-class Lighthill-Whitham-Richards (MCLWR) traffic flow model. This scheme is based on high-order reconstruction for spatial discretization and an implicit-explicit Runge-Kutta method for time integration. The resulting method retains the simplicity of the relaxation schemes. There is no need to involve Riemann solvers and characteristic decomposition. Even the computation of the eigenvalues is not required. This makes the scheme particularly well suited for the MCLWR model in which the analytical expressions of the eigenvalues are difficult to obtain for more than four classes of road users. The numerical results illustrate the effectiveness of the presented method.
基金University Science Foundation of Jiangsu Province (04KJD510018)
文摘Hierarchical Support Vector Machine （H-SVM） is faster in training and classification than other usual multi-class SVMs such as ＂1-V-R＂and ＂1-V-1＂. In this paper, a new multi-class fault diagnosis algorithm based on H-SVM is proposed and applied to aero-engine. Before SVM training, the training data are first clustered according to their class-center Euclid distances in some feature spaces. The samples which have close distances are divided into the same sub-classes for training, and this makes the H-SVM have reasonable hierarchical construction and good generalization performance. Instead of the common C-SVM, the v-SVM is selected as the binary classifier, in which the parameter v varies only from 0 to 1 and can be determined more easily. The simulation results show that the designed H-SVMs can fast diagnose the multi-class single faults and combination faults for the gas path components of an aero-engine. The fault classifiers have good diagnosis accuracy and can keep robust even when the measurement inputs are disturbed by noises.
文摘Multi-source multi-class classification methods based on multi-class Support Vector Machines and data fusion strategies are proposed in this paper. The centralized and distributed fusion schemes are applied to combine information from several data sources. In the centralized scheme, all information from several data sources is centralized to construct an input space. Then a multi-class Support Vector Machine classifier is trained. In the distributed schemes, the individual data sources are proc-essed separately and modelled by using the multi-class Support Vector Machine. Then new data fusion strategies are proposed to combine the information from the individual multi-class Support Vector Machine models. Our proposed fusion strategies take into account that an Support Vector Machine (SVM) classifier achieves classification by finding the optimal classification hyperplane with maximal margin. The proposed methods are applied for fault diagnosis of a diesel engine. The experimental results showed that almost all the proposed approaches can largely improve the diagnostic accuracy. The robustness of diagnosis is also improved because of the implementation of data fusion strategies. The proposed methods can also be applied in other fields.
