The complex sand-casting process combined with the interactions between process parameters makes it difficult to control the casting quality,resulting in a high scrap rate.A strategy based on a data-driven model was p...The complex sand-casting process combined with the interactions between process parameters makes it difficult to control the casting quality,resulting in a high scrap rate.A strategy based on a data-driven model was proposed to reduce casting defects and improve production efficiency,which includes the random forest(RF)classification model,the feature importance analysis,and the process parameters optimization with Monte Carlo simulation.The collected data includes four types of defects and corresponding process parameters were used to construct the RF model.Classification results show a recall rate above 90% for all categories.The Gini Index was used to assess the importance of the process parameters in the formation of various defects in the RF model.Finally,the classification model was applied to different production conditions for quality prediction.In the case of process parameters optimization for gas porosity defects,this model serves as an experimental process in the Monte Carlo method to estimate a better temperature distribution.The prediction model,when applied to the factory,greatly improved the efficiency of defect detection.Results show that the scrap rate decreased from 10.16% to 6.68%.展开更多
Complex proteins are needed for many biological activities.Folding amino acid chains reveals their properties and functions.They support healthy tissue structure,physiology,and homeostasis.Precision medicine and treat...Complex proteins are needed for many biological activities.Folding amino acid chains reveals their properties and functions.They support healthy tissue structure,physiology,and homeostasis.Precision medicine and treatments require quantitative protein identification and function.Despite technical advances and protein sequence data exploration,bioinformatics’“basic structure”problem—the automatic deduction of a protein’s properties from its amino acid sequence—remains unsolved.Protein function inference from amino acid sequences is the main biological data challenge.This study analyzes whether raw sequencing can characterize biological facts.A massive corpus of protein sequences and the Globin-like superfamily’s related protein families generate a solid vector representation.A coding technique for each sequence in each family was devised using two representations to identify each amino acid precisely.A bispectral analysis converts encoded protein numerical sequences into images for better protein sequence and family discrimination.Training and validation employed 70%of the dataset,while 30%was used for testing.This paper examined the performance of multistage deep learning models for differentiating between sixteen protein families after encoding and representing each encoded sequence by a higher spectral representation image(Bispectrum).Cascading minimized false positive and negative cases in all phases.The initial stage focused on two classes(six groups and ten groups).The subsequent stages focused on the few classes almost accurately separated in the first stage and decreased the overlapping cases between families that appeared in single-stage deep learning classification.The single-stage technique had 64.2%+/-22.8%accuracy,63.3%+/-17.1%precision,and a 63.2%+/19.4%F1-score.The two-stage technique yielded 92.2%+/-4.9%accuracy,92.7%+/-7.0%precision,and a 92.3%+/-5.0%F1-score.This work provides balanced,reliable,and precise forecasts for all families in all measures.It ensured that the new model was resilient to family variances and provided high-scoring results.展开更多
Proteins are essential for many biological functions.For example,folding amino acid chains reveals their functionalities by maintaining tissue structure,physiology,and homeostasis.Note that quantifiable protein charac...Proteins are essential for many biological functions.For example,folding amino acid chains reveals their functionalities by maintaining tissue structure,physiology,and homeostasis.Note that quantifiable protein characteristics are vital for improving therapies and precision medicine.The automatic inference of a protein’s properties from its amino acid sequence is called“basic structure”.Nevertheless,it remains a critical unsolved challenge in bioinformatics,although with recent technological advances and the investigation of protein sequence data.Inferring protein function from amino acid sequences is crucial in biology.This study considers using raw sequencing to explain biological facts using a large corpus of protein sequences and the Globin-like superfamily to generate a vector representation.The power of two representations was used to identify each amino acid,and a coding technique was established for each sequence family.Subsequently,the encoded protein numerical sequences are transformed into an image using bispectral analysis to identify essential characteristics for discriminating between protein sequences and their families.A deep Convolutional Neural Network(CNN)classifies the resulting images and developed non-normalized and normalized encoding techniques.Initially,the dataset was split 70/30 for training and testing.Correspondingly,the dataset was utilized for 70%training,15%validation,and 15%testing.The suggested methods are evaluated using accuracy,precision,and recall.The non-normalized method had 70%accuracy,72%precision,and 71%recall.68%accuracy,67%precision,and 67%recall after validation.Meanwhile,the normalized approach without validation had 92.4%accuracy,94.3%precision,and 91.1%recall.Validation showed 90%accuracy,91.2%precision,and 89.7%recall.Note that both algorithms outperform the rest.The paper presents that bispectrum-based nonlinear analysis using deep learning models outperforms standard machine learning methods and other deep learning methods based on convolutional architecture.They offered the best inference performance as the proposed approach improves categorization and prediction.Several instances show successful multi-class prediction in molecular biology’s massive data.展开更多
Question-answering(QA)models find answers to a given question.The necessity of automatically finding answers is increasing because it is very important and challenging from the large-scale QA data sets.In this paper,w...Question-answering(QA)models find answers to a given question.The necessity of automatically finding answers is increasing because it is very important and challenging from the large-scale QA data sets.In this paper,we deal with the QA pair matching approach in QA models,which finds the most relevant question and its recommended answer for a given question.Existing studies for the approach performed on the entire dataset or datasets within a category that the question writer manually specifies.In contrast,we aim to automatically find the category to which the question belongs by employing the text classification model and to find the answer corresponding to the question within the category.Due to the text classification model,we can effectively reduce the search space for finding the answers to a given question.Therefore,the proposed model improves the accuracy of the QA matching model and significantly reduces the model inference time.Furthermore,to improve the performance of finding similar sentences in each category,we present an ensemble embedding model for sentences,improving the performance compared to the individual embedding models.Using real-world QA data sets,we evaluate the performance of the proposed QA matching model.As a result,the accuracy of our final ensemble embedding model based on the text classification model is 81.18%,which outperforms the existing models by 9.81%∼14.16%point.Moreover,in terms of the model inference speed,our model is faster than the existing models by 2.61∼5.07 times due to the effective reduction of search spaces by the text classification model.展开更多
This paper aims to build an employee attrition classification model based on the Stacking algorithm.Oversampling algorithm is applied to address the issue of data imbalance and the Randomforest feature importance rank...This paper aims to build an employee attrition classification model based on the Stacking algorithm.Oversampling algorithm is applied to address the issue of data imbalance and the Randomforest feature importance ranking method is used to resolve the overfitting problem after data cleaning and preprocessing.Then,different algorithms are used to establish classification models as control experiments,and R-squared indicators are used to compare.Finally,the Stacking algorithm is used to establish the final classification model.This model has practical and significant implications for both human resource management and employee attrition analysis.展开更多
Deep Neural Networks(DNNs)are integral to various aspects of modern life,enhancing work efficiency.Nonethe-less,their susceptibility to diverse attack methods,including backdoor attacks,raises security concerns.We aim...Deep Neural Networks(DNNs)are integral to various aspects of modern life,enhancing work efficiency.Nonethe-less,their susceptibility to diverse attack methods,including backdoor attacks,raises security concerns.We aim to investigate backdoor attack methods for image categorization tasks,to promote the development of DNN towards higher security.Research on backdoor attacks currently faces significant challenges due to the distinct and abnormal data patterns of malicious samples,and the meticulous data screening by developers,hindering practical attack implementation.To overcome these challenges,this study proposes a Gaussian Noise-Targeted Universal Adversarial Perturbation(GN-TUAP)algorithm.This approach restricts the direction of perturbations and normalizes abnormal pixel values,ensuring that perturbations progress as much as possible in a direction perpendicular to the decision hyperplane in linear problems.This limits anomalies within the perturbations improves their visual stealthiness,and makes them more challenging for defense methods to detect.To verify the effectiveness,stealthiness,and robustness of GN-TUAP,we proposed a comprehensive threat model.Based on this model,extensive experiments were conducted using the CIFAR-10,CIFAR-100,GTSRB,and MNIST datasets,comparing our method with existing state-of-the-art attack methods.We also tested our perturbation triggers using various defense methods and further experimented on the robustness of the triggers against noise filtering techniques.The experimental outcomes demonstrate that backdoor attacks leveraging perturbations generated via our algorithm exhibit cross-model attack effectiveness and superior stealthiness.Furthermore,they possess robust anti-detection capabilities and maintain commendable performance when subjected to noise-filtering methods.展开更多
Accurate simulation of tropical cyclone tracks is a prerequisite for tropical cyclone risk assessment.Against the spatial characteristics of tropical cyclone tracks in the Northwest Pacific region,stochastic simulatio...Accurate simulation of tropical cyclone tracks is a prerequisite for tropical cyclone risk assessment.Against the spatial characteristics of tropical cyclone tracks in the Northwest Pacific region,stochastic simulation method based on classification model is used to simulate tropical cyclone tracks in this region.Such simulation includes the classification method,the genesis model,the traveling model,and the lysis model.Tropical cyclone tracks in the Northwest Pacific region are classified into five categories on the basis of its movement characteristics and steering positions.In the genesis model,Gaussian kernel probability density functions with the biased cross validation method are used to simulate the annual occurrence number and genesis positions.The traveling model is established on the basis of the mean and mean square error of the historical 6 h latitude and longitude displacements.The termination probability is used as the discrimination standard in the lysis model.Then,this stochastic simulation method of tropical cyclone tracks is applied and qualitatively evaluated with different diagnostics.Results show that the tropical cyclone tracks in Northwest Pacific can be satisfactorily simulated with this classification model.展开更多
Objective Debris flows are cohesive sediment gravity flows which occur in both subaerial and subaqueous settings. Compared to subaerial debris flows which have been well studied as a geological hazard, subaqueous deb...Objective Debris flows are cohesive sediment gravity flows which occur in both subaerial and subaqueous settings. Compared to subaerial debris flows which have been well studied as a geological hazard, subaqueous debris flows showing complicated sediment composition and sedimentary processes were poorly understood. The main objective of this work is to establish a classification scheme and facies sequence models of subaqueous debris flows for well understanding their sedimentary processes and depositional characteristics.展开更多
Prognosis is a key technology to improve reliability,safety and maintainability of products,a lot of researchers have been devoted to this technology.But to improve the predict accuracy of remaining life of products h...Prognosis is a key technology to improve reliability,safety and maintainability of products,a lot of researchers have been devoted to this technology.But to improve the predict accuracy of remaining life of products has been difficult.To predict the lifetime specification of pneumatic cylinders with high reliability and long lifetime and small specimen,this paper put forward the prognosis algorithm based on the path classification and estimation(PACE) model.PACE model is based entirely on failure data instead of failure threshold.Pneumatic cylinders normally characterize with failure mechanism wear and tear.Since the minimum working pressure increases with the number of working cycles,the minimum working pressure is chosen as degradation signal.PACE model is fundamentally composed of two operations:path classification and remaining useful life(RUL) estimation.Path classification is to classify a current degradation path as belonging to one or more of previously collected exemplary degradation paths.RUL estimation is to use the resulting memberships to estimate the remaining useful life.In order for verification and validation of PACE prognostic method,six pneumatic cylinders are tested.The test data is analyzed by PACE prognostics.It is found that the PACE based prognosis method has higher prediction accuracy and smaller variance and PACE model is significantly outperform population based prognostics especially for small specimen condition.PACE model based method solved the problem of prediction accuracy for small specimen pneumatic cylinders' prognosis.展开更多
There has been evidence of crime in the US since colonization. In this article, we analyze the crime statistics of San Francisco and its resolution of crime recorded from January to September of the year 2018. We defi...There has been evidence of crime in the US since colonization. In this article, we analyze the crime statistics of San Francisco and its resolution of crime recorded from January to September of the year 2018. We define resolution of crime as a target variable and study its relationship with other variables. We make several classification models to predict resolution of crime using several data mining techniques and suggest the best model for predicting resolution.展开更多
A new classification model for host intrusion detection based on the unidentified short sequences and RIPPER algorithm is proposed. The concepts of different short sequences on the system call traces are strictly defi...A new classification model for host intrusion detection based on the unidentified short sequences and RIPPER algorithm is proposed. The concepts of different short sequences on the system call traces are strictly defined on the basis of in-depth analysis of completeness and correctness of pattern databases. Labels of short sequences are predicted by learned RIPPER rule set and the nature of the unidentified short sequences is confirmed by statistical method. Experiment results indicate that the classification model increases clearly the deviation between the attack and the normal traces and improves detection capability against known and unknown attacks.展开更多
In agriculture,rice plant disease diagnosis has become a challenging issue,and early identification of this disease can avoid huge loss incurred from less crop productivity.Some of the recently-developed computer visi...In agriculture,rice plant disease diagnosis has become a challenging issue,and early identification of this disease can avoid huge loss incurred from less crop productivity.Some of the recently-developed computer vision and Deep Learning(DL)approaches can be commonly employed in designing effective models for rice plant disease detection and classification processes.With this motivation,the current research work devises an Efficient Deep Learning based FusionModel for Rice Plant Disease(EDLFM-RPD)detection and classification.The aim of the proposed EDLFM-RPD technique is to detect and classify different kinds of rice plant diseases in a proficient manner.In addition,EDLFM-RPD technique involves median filtering-based preprocessing and K-means segmentation to determine the infected portions.The study also used a fusion of handcrafted Gray Level Co-occurrence Matrix(GLCM)and Inception-based deep features to derive the features.Finally,Salp Swarm Optimization with Fuzzy Support Vector Machine(FSVM)model is utilized for classification.In order to validate the enhanced outcomes of EDLFM-RPD technique,a series of simulations was conducted.The results were assessed under different measures.The obtained values infer the improved performance of EDLFM-RPD technique over recent approaches and achieved a maximum accuracy of 96.170%.展开更多
A model classification rate control method for video coding is proposed. The macro-blocks are classified according to their prediction errors, and different parameters are used in the rate-quantization and distortion-...A model classification rate control method for video coding is proposed. The macro-blocks are classified according to their prediction errors, and different parameters are used in the rate-quantization and distortion-quantization model. The different model parameters are calculated from the previous frame of the same type in the process of coding. These models are used to estimate the relations among rate, distortion and quantization of the current frame. Further steps, such as R-D optimization based quantization adjustment and smoothing of quantization of adjacent macrobloeks, are used to improve the quality. The results of the experiments prove that the technique is effective and can be realized easily. The method presented in the paper can be a good way for MPEG and H. 264 rate control.展开更多
In this paper,we consider the statistical inference problems for the fixed effect and variance component functions in the two-way classification random effects model with skewnormal errors.Firstly,the exact test stati...In this paper,we consider the statistical inference problems for the fixed effect and variance component functions in the two-way classification random effects model with skewnormal errors.Firstly,the exact test statistic for the fixed effect is constructed.Secondly,using the Bootstrap approach and generalized approach,the one-sided hypothesis testing and interval estimation problems for the single variance component,the sum and ratio of variance components are discussed respectively.Further,the Monte Carlo simulation results indicate that the exact test statistic performs well in the one-sided hypothesis testing problem for the fixed effect.And the Bootstrap approach is better than the generalized approach in the one-sided hypothesis testing problems for variance component functions in most cases.Finally,the above approaches are applied to the real data examples of the consumer price index and value-added index of three industries to verify their rationality and effectiveness.展开更多
In this paper, we explore the multi-classification problem of acupuncture acupoints bas</span><span><span style="font-family:Verdana;">ed on </span><span style="font-family:Ve...In this paper, we explore the multi-classification problem of acupuncture acupoints bas</span><span><span style="font-family:Verdana;">ed on </span><span style="font-family:Verdana;">Bert</span><span style="font-family:Verdana;"> model, </span><i><span style="font-family:Verdana;">i.e.</span></i><span style="font-family:Verdana;">, we try to recommend the best main acupuncture point for treating the disease by classifying and predicting the main acupuncture point for the disease, and further explore its acupuncture point grouping to provide the medical practitioner with the optimal solution for treating the disease and improv</span></span></span><span style="font-family:Verdana;">ing</span><span style="font-family:""><span style="font-family:Verdana;"> the clinical decision-making ability. The Bert-Chinese-Acupoint model was constructed by retraining </span><span style="font-family:Verdana;">on the basis of</span><span style="font-family:Verdana;"> the Bert model, and the semantic features in terms of acupuncture points were added to the acupunctu</span></span><span style="font-family:""><span style="font-family:Verdana;">re point corpus in the fine-tuning process to increase the semantic features in terms of acupuncture </span><span style="font-family:Verdana;">points,</span><span style="font-family:Verdana;"> and compared with the machine learning method. The results show that the Bert-Chinese Acupoint model proposed in this paper has a 3% improvement in accuracy compared to the </span><span style="font-family:Verdana;">best performing</span><span style="font-family:Verdana;"> model in the machine learning approach.展开更多
Nowadays,the usage of socialmedia platforms is rapidly increasing,and rumours or false information are also rising,especially among Arab nations.This false information is harmful to society and individuals.Blocking an...Nowadays,the usage of socialmedia platforms is rapidly increasing,and rumours or false information are also rising,especially among Arab nations.This false information is harmful to society and individuals.Blocking and detecting the spread of fake news in Arabic becomes critical.Several artificial intelligence(AI)methods,including contemporary transformer techniques,BERT,were used to detect fake news.Thus,fake news in Arabic is identified by utilizing AI approaches.This article develops a new hunterprey optimization with hybrid deep learning-based fake news detection(HPOHDL-FND)model on the Arabic corpus.The HPOHDL-FND technique undergoes extensive data pre-processing steps to transform the input data into a useful format.Besides,the HPOHDL-FND technique utilizes long-term memory with a recurrent neural network(LSTM-RNN)model for fake news detection and classification.Finally,hunter prey optimization(HPO)algorithm is exploited for optimal modification of the hyperparameters related to the LSTM-RNN model.The performance validation of the HPOHDL-FND technique is tested using two Arabic datasets.The outcomes exemplified better performance over the other existing techniques with maximum accuracy of 96.57%and 93.53%on Covid19Fakes and satirical datasets,respectively.展开更多
Neural networks play a significant role in the field of image classification.When an input image is modified by adversarial attacks,the changes are imperceptible to the human eye,but it still leads to misclassificatio...Neural networks play a significant role in the field of image classification.When an input image is modified by adversarial attacks,the changes are imperceptible to the human eye,but it still leads to misclassification of the images.Researchers have demonstrated these attacks to make production self-driving cars misclassify StopRoad signs as 45 Miles Per Hour(MPH)road signs and a turtle being misclassified as AK47.Three primary types of defense approaches exist which can safeguard against such attacks i.e.,Gradient Masking,Robust Optimization,and Adversarial Example Detection.Very few approaches use Generative Adversarial Networks(GAN)for Defense against Adversarial Attacks.In this paper,we create a new approach to defend against adversarial attacks,dubbed Chained Dual-Generative Adversarial Network(CD-GAN)that tackles the defense against adversarial attacks by minimizing the perturbations of the adversarial image using iterative oversampling and undersampling using GANs.CD-GAN is created using two GANs,i.e.,CDGAN’s Sub-ResolutionGANandCDGAN’s Super-ResolutionGAN.The first is CDGAN’s Sub-Resolution GAN which takes the original resolution input image and oversamples it to generate a lower resolution neutralized image.The second is CDGAN’s Super-Resolution GAN which takes the output of the CDGAN’s Sub-Resolution and undersamples,it to generate the higher resolution image which removes any remaining perturbations.Chained Dual GAN is formed by chaining these two GANs together.Both of these GANs are trained independently.CDGAN’s Sub-Resolution GAN is trained using higher resolution adversarial images as inputs and lower resolution neutralized images as output image examples.Hence,this GAN downscales the image while removing adversarial attack noise.CDGAN’s Super-Resolution GAN is trained using lower resolution adversarial images as inputs and higher resolution neutralized images as output images.Because of this,it acts as an Upscaling GAN while removing the adversarial attak noise.Furthermore,CD-GAN has a modular design such that it can be prefixed to any existing classifier without any retraining or extra effort,and 2542 CMC,2023,vol.74,no.2 can defend any classifier model against adversarial attack.In this way,it is a Generalized Defense against adversarial attacks,capable of defending any classifier model against any attacks.This enables the user to directly integrate CD-GANwith an existing production deployed classifier smoothly.CD-GAN iteratively removes the adversarial noise using a multi-step approach in a modular approach.It performs comparably to the state of the arts with mean accuracy of 33.67 while using minimal compute resources in training.展开更多
Since the late 2010s,Artificial Intelligence(AI)including machine learning,boosted through deep learning,has boomed as a vital tool to leverage computer vision,natural language processing and speech recognition in rev...Since the late 2010s,Artificial Intelligence(AI)including machine learning,boosted through deep learning,has boomed as a vital tool to leverage computer vision,natural language processing and speech recognition in revolutionizing zoological research.This review provides an overview of the primary tasks,core models,datasets,and applications of AI in zoological research,including animal classification,resource conservation,behavior,development,genetics and evolution,breeding and health,disease models,and paleontology.Additionally,we explore the challenges and future directions of integrating AI into this field.Based on numerous case studies,this review outlines various avenues for incorporating AI into zoological research and underscores its potential to enhance our understanding of the intricate relationships that exist within the animal kingdom.As we build a bridge between beast and byte realms,this review serves as a resource for envisioning novel AI applications in zoological research that have not yet been explored.展开更多
Cardiac diseases are one of the greatest global health challenges.Due to the high annual mortality rates,cardiac diseases have attracted the attention of numerous researchers in recent years.This article proposes a hy...Cardiac diseases are one of the greatest global health challenges.Due to the high annual mortality rates,cardiac diseases have attracted the attention of numerous researchers in recent years.This article proposes a hybrid fuzzy fusion classification model for cardiac arrhythmia diseases.The fusion model is utilized to optimally select the highest-ranked features generated by a variety of well-known feature-selection algorithms.An ensemble of classifiers is then applied to the fusion’s results.The proposed model classifies the arrhythmia dataset from the University of California,Irvine into normal/abnormal classes as well as 16 classes of arrhythmia.Initially,at the preprocessing steps,for the miss-valued attributes,we used the average value in the linear attributes group by the same class and the most frequent value for nominal attributes.However,in order to ensure the model optimality,we eliminated all attributes which have zero or constant values that might bias the results of utilized classifiers.The preprocessing step led to 161 out of 279 attributes(features).Thereafter,a fuzzy-based feature-selection fusion method is applied to fuse high-ranked features obtained from different heuristic feature-selection algorithms.In short,our study comprises three main blocks:(1)sensing data and preprocessing;(2)feature queuing,selection,and extraction;and(3)the predictive model.Our proposed method improves classification performance in terms of accuracy,F1measure,recall,and precision when compared to state-of-the-art techniques.It achieves 98.5%accuracy for binary class mode and 98.9%accuracy for categorized class mode.展开更多
Glass is the precious material evidence of the trade of the early Silk Road. The ancient glass was easily affected by the environmental impact and weathering, and the change of composition ratios affected the correct ...Glass is the precious material evidence of the trade of the early Silk Road. The ancient glass was easily affected by the environmental impact and weathering, and the change of composition ratios affected the correct judgment of its category. In this paper, mathematical models and methods such as Chi-square test, weighted average method, principal component analysis, cluster analysis, binary classification model and grey correlation analysis were used comprehensively to analyze the data of sample glass products combined with their categories. The results showed that the weathered high-potassium glass could be divided into 12, 9, 10 and 27, 7, 22 and so on.展开更多
基金financially supported by the National Key Research and Development Program of China(2022YFB3706800,2020YFB1710100)the National Natural Science Foundation of China(51821001,52090042,52074183)。
文摘The complex sand-casting process combined with the interactions between process parameters makes it difficult to control the casting quality,resulting in a high scrap rate.A strategy based on a data-driven model was proposed to reduce casting defects and improve production efficiency,which includes the random forest(RF)classification model,the feature importance analysis,and the process parameters optimization with Monte Carlo simulation.The collected data includes four types of defects and corresponding process parameters were used to construct the RF model.Classification results show a recall rate above 90% for all categories.The Gini Index was used to assess the importance of the process parameters in the formation of various defects in the RF model.Finally,the classification model was applied to different production conditions for quality prediction.In the case of process parameters optimization for gas porosity defects,this model serves as an experimental process in the Monte Carlo method to estimate a better temperature distribution.The prediction model,when applied to the factory,greatly improved the efficiency of defect detection.Results show that the scrap rate decreased from 10.16% to 6.68%.
文摘Complex proteins are needed for many biological activities.Folding amino acid chains reveals their properties and functions.They support healthy tissue structure,physiology,and homeostasis.Precision medicine and treatments require quantitative protein identification and function.Despite technical advances and protein sequence data exploration,bioinformatics’“basic structure”problem—the automatic deduction of a protein’s properties from its amino acid sequence—remains unsolved.Protein function inference from amino acid sequences is the main biological data challenge.This study analyzes whether raw sequencing can characterize biological facts.A massive corpus of protein sequences and the Globin-like superfamily’s related protein families generate a solid vector representation.A coding technique for each sequence in each family was devised using two representations to identify each amino acid precisely.A bispectral analysis converts encoded protein numerical sequences into images for better protein sequence and family discrimination.Training and validation employed 70%of the dataset,while 30%was used for testing.This paper examined the performance of multistage deep learning models for differentiating between sixteen protein families after encoding and representing each encoded sequence by a higher spectral representation image(Bispectrum).Cascading minimized false positive and negative cases in all phases.The initial stage focused on two classes(six groups and ten groups).The subsequent stages focused on the few classes almost accurately separated in the first stage and decreased the overlapping cases between families that appeared in single-stage deep learning classification.The single-stage technique had 64.2%+/-22.8%accuracy,63.3%+/-17.1%precision,and a 63.2%+/19.4%F1-score.The two-stage technique yielded 92.2%+/-4.9%accuracy,92.7%+/-7.0%precision,and a 92.3%+/-5.0%F1-score.This work provides balanced,reliable,and precise forecasts for all families in all measures.It ensured that the new model was resilient to family variances and provided high-scoring results.
文摘Proteins are essential for many biological functions.For example,folding amino acid chains reveals their functionalities by maintaining tissue structure,physiology,and homeostasis.Note that quantifiable protein characteristics are vital for improving therapies and precision medicine.The automatic inference of a protein’s properties from its amino acid sequence is called“basic structure”.Nevertheless,it remains a critical unsolved challenge in bioinformatics,although with recent technological advances and the investigation of protein sequence data.Inferring protein function from amino acid sequences is crucial in biology.This study considers using raw sequencing to explain biological facts using a large corpus of protein sequences and the Globin-like superfamily to generate a vector representation.The power of two representations was used to identify each amino acid,and a coding technique was established for each sequence family.Subsequently,the encoded protein numerical sequences are transformed into an image using bispectral analysis to identify essential characteristics for discriminating between protein sequences and their families.A deep Convolutional Neural Network(CNN)classifies the resulting images and developed non-normalized and normalized encoding techniques.Initially,the dataset was split 70/30 for training and testing.Correspondingly,the dataset was utilized for 70%training,15%validation,and 15%testing.The suggested methods are evaluated using accuracy,precision,and recall.The non-normalized method had 70%accuracy,72%precision,and 71%recall.68%accuracy,67%precision,and 67%recall after validation.Meanwhile,the normalized approach without validation had 92.4%accuracy,94.3%precision,and 91.1%recall.Validation showed 90%accuracy,91.2%precision,and 89.7%recall.Note that both algorithms outperform the rest.The paper presents that bispectrum-based nonlinear analysis using deep learning models outperforms standard machine learning methods and other deep learning methods based on convolutional architecture.They offered the best inference performance as the proposed approach improves categorization and prediction.Several instances show successful multi-class prediction in molecular biology’s massive data.
基金This work was supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.2022R1F1A1067008)by the Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(No.2019R1A6A1A03032119).
文摘Question-answering(QA)models find answers to a given question.The necessity of automatically finding answers is increasing because it is very important and challenging from the large-scale QA data sets.In this paper,we deal with the QA pair matching approach in QA models,which finds the most relevant question and its recommended answer for a given question.Existing studies for the approach performed on the entire dataset or datasets within a category that the question writer manually specifies.In contrast,we aim to automatically find the category to which the question belongs by employing the text classification model and to find the answer corresponding to the question within the category.Due to the text classification model,we can effectively reduce the search space for finding the answers to a given question.Therefore,the proposed model improves the accuracy of the QA matching model and significantly reduces the model inference time.Furthermore,to improve the performance of finding similar sentences in each category,we present an ensemble embedding model for sentences,improving the performance compared to the individual embedding models.Using real-world QA data sets,we evaluate the performance of the proposed QA matching model.As a result,the accuracy of our final ensemble embedding model based on the text classification model is 81.18%,which outperforms the existing models by 9.81%∼14.16%point.Moreover,in terms of the model inference speed,our model is faster than the existing models by 2.61∼5.07 times due to the effective reduction of search spaces by the text classification model.
文摘This paper aims to build an employee attrition classification model based on the Stacking algorithm.Oversampling algorithm is applied to address the issue of data imbalance and the Randomforest feature importance ranking method is used to resolve the overfitting problem after data cleaning and preprocessing.Then,different algorithms are used to establish classification models as control experiments,and R-squared indicators are used to compare.Finally,the Stacking algorithm is used to establish the final classification model.This model has practical and significant implications for both human resource management and employee attrition analysis.
基金funded by National Natural Science Foundation of China under Grant No.61806171The Sichuan University of Science&Engineering Talent Project under Grant No.2021RC15Sichuan University of Science&Engineering Graduate Student Innovation Fund under Grant No.Y2023115,The Scientific Research and Innovation Team Program of Sichuan University of Science and Technology under Grant No.SUSE652A006.
文摘Deep Neural Networks(DNNs)are integral to various aspects of modern life,enhancing work efficiency.Nonethe-less,their susceptibility to diverse attack methods,including backdoor attacks,raises security concerns.We aim to investigate backdoor attack methods for image categorization tasks,to promote the development of DNN towards higher security.Research on backdoor attacks currently faces significant challenges due to the distinct and abnormal data patterns of malicious samples,and the meticulous data screening by developers,hindering practical attack implementation.To overcome these challenges,this study proposes a Gaussian Noise-Targeted Universal Adversarial Perturbation(GN-TUAP)algorithm.This approach restricts the direction of perturbations and normalizes abnormal pixel values,ensuring that perturbations progress as much as possible in a direction perpendicular to the decision hyperplane in linear problems.This limits anomalies within the perturbations improves their visual stealthiness,and makes them more challenging for defense methods to detect.To verify the effectiveness,stealthiness,and robustness of GN-TUAP,we proposed a comprehensive threat model.Based on this model,extensive experiments were conducted using the CIFAR-10,CIFAR-100,GTSRB,and MNIST datasets,comparing our method with existing state-of-the-art attack methods.We also tested our perturbation triggers using various defense methods and further experimented on the robustness of the triggers against noise filtering techniques.The experimental outcomes demonstrate that backdoor attacks leveraging perturbations generated via our algorithm exhibit cross-model attack effectiveness and superior stealthiness.Furthermore,they possess robust anti-detection capabilities and maintain commendable performance when subjected to noise-filtering methods.
基金National Natural Science Foundation of China(51408174)Provincial Undergraduate Innovation and Entrepreneurship Training Program of Hefei University of Technology(S201910359302)
文摘Accurate simulation of tropical cyclone tracks is a prerequisite for tropical cyclone risk assessment.Against the spatial characteristics of tropical cyclone tracks in the Northwest Pacific region,stochastic simulation method based on classification model is used to simulate tropical cyclone tracks in this region.Such simulation includes the classification method,the genesis model,the traveling model,and the lysis model.Tropical cyclone tracks in the Northwest Pacific region are classified into five categories on the basis of its movement characteristics and steering positions.In the genesis model,Gaussian kernel probability density functions with the biased cross validation method are used to simulate the annual occurrence number and genesis positions.The traveling model is established on the basis of the mean and mean square error of the historical 6 h latitude and longitude displacements.The termination probability is used as the discrimination standard in the lysis model.Then,this stochastic simulation method of tropical cyclone tracks is applied and qualitatively evaluated with different diagnostics.Results show that the tropical cyclone tracks in Northwest Pacific can be satisfactorily simulated with this classification model.
基金jointly funded by the National Natural Science Foundation of China(grants No.41172104,41202078 and 41372117)the Major National S&T Program of China(grant No.2011ZX05009-002)
文摘Objective Debris flows are cohesive sediment gravity flows which occur in both subaerial and subaqueous settings. Compared to subaerial debris flows which have been well studied as a geological hazard, subaqueous debris flows showing complicated sediment composition and sedimentary processes were poorly understood. The main objective of this work is to establish a classification scheme and facies sequence models of subaqueous debris flows for well understanding their sedimentary processes and depositional characteristics.
基金supported by the Laboratory of Aviation Safety Technical Analysis and Appraisal of China Academy of Civil Aviation Science and Technology(Grant No. 2009-02)
文摘Prognosis is a key technology to improve reliability,safety and maintainability of products,a lot of researchers have been devoted to this technology.But to improve the predict accuracy of remaining life of products has been difficult.To predict the lifetime specification of pneumatic cylinders with high reliability and long lifetime and small specimen,this paper put forward the prognosis algorithm based on the path classification and estimation(PACE) model.PACE model is based entirely on failure data instead of failure threshold.Pneumatic cylinders normally characterize with failure mechanism wear and tear.Since the minimum working pressure increases with the number of working cycles,the minimum working pressure is chosen as degradation signal.PACE model is fundamentally composed of two operations:path classification and remaining useful life(RUL) estimation.Path classification is to classify a current degradation path as belonging to one or more of previously collected exemplary degradation paths.RUL estimation is to use the resulting memberships to estimate the remaining useful life.In order for verification and validation of PACE prognostic method,six pneumatic cylinders are tested.The test data is analyzed by PACE prognostics.It is found that the PACE based prognosis method has higher prediction accuracy and smaller variance and PACE model is significantly outperform population based prognostics especially for small specimen condition.PACE model based method solved the problem of prediction accuracy for small specimen pneumatic cylinders' prognosis.
文摘There has been evidence of crime in the US since colonization. In this article, we analyze the crime statistics of San Francisco and its resolution of crime recorded from January to September of the year 2018. We define resolution of crime as a target variable and study its relationship with other variables. We make several classification models to predict resolution of crime using several data mining techniques and suggest the best model for predicting resolution.
文摘A new classification model for host intrusion detection based on the unidentified short sequences and RIPPER algorithm is proposed. The concepts of different short sequences on the system call traces are strictly defined on the basis of in-depth analysis of completeness and correctness of pattern databases. Labels of short sequences are predicted by learned RIPPER rule set and the nature of the unidentified short sequences is confirmed by statistical method. Experiment results indicate that the classification model increases clearly the deviation between the attack and the normal traces and improves detection capability against known and unknown attacks.
文摘In agriculture,rice plant disease diagnosis has become a challenging issue,and early identification of this disease can avoid huge loss incurred from less crop productivity.Some of the recently-developed computer vision and Deep Learning(DL)approaches can be commonly employed in designing effective models for rice plant disease detection and classification processes.With this motivation,the current research work devises an Efficient Deep Learning based FusionModel for Rice Plant Disease(EDLFM-RPD)detection and classification.The aim of the proposed EDLFM-RPD technique is to detect and classify different kinds of rice plant diseases in a proficient manner.In addition,EDLFM-RPD technique involves median filtering-based preprocessing and K-means segmentation to determine the infected portions.The study also used a fusion of handcrafted Gray Level Co-occurrence Matrix(GLCM)and Inception-based deep features to derive the features.Finally,Salp Swarm Optimization with Fuzzy Support Vector Machine(FSVM)model is utilized for classification.In order to validate the enhanced outcomes of EDLFM-RPD technique,a series of simulations was conducted.The results were assessed under different measures.The obtained values infer the improved performance of EDLFM-RPD technique over recent approaches and achieved a maximum accuracy of 96.170%.
基金This project was supported by the High Technology Research and Development Programof China (2002AA103087) .
文摘A model classification rate control method for video coding is proposed. The macro-blocks are classified according to their prediction errors, and different parameters are used in the rate-quantization and distortion-quantization model. The different model parameters are calculated from the previous frame of the same type in the process of coding. These models are used to estimate the relations among rate, distortion and quantization of the current frame. Further steps, such as R-D optimization based quantization adjustment and smoothing of quantization of adjacent macrobloeks, are used to improve the quality. The results of the experiments prove that the technique is effective and can be realized easily. The method presented in the paper can be a good way for MPEG and H. 264 rate control.
基金supported by National Social Science Foundation of China(21BTJ068)。
文摘In this paper,we consider the statistical inference problems for the fixed effect and variance component functions in the two-way classification random effects model with skewnormal errors.Firstly,the exact test statistic for the fixed effect is constructed.Secondly,using the Bootstrap approach and generalized approach,the one-sided hypothesis testing and interval estimation problems for the single variance component,the sum and ratio of variance components are discussed respectively.Further,the Monte Carlo simulation results indicate that the exact test statistic performs well in the one-sided hypothesis testing problem for the fixed effect.And the Bootstrap approach is better than the generalized approach in the one-sided hypothesis testing problems for variance component functions in most cases.Finally,the above approaches are applied to the real data examples of the consumer price index and value-added index of three industries to verify their rationality and effectiveness.
文摘In this paper, we explore the multi-classification problem of acupuncture acupoints bas</span><span><span style="font-family:Verdana;">ed on </span><span style="font-family:Verdana;">Bert</span><span style="font-family:Verdana;"> model, </span><i><span style="font-family:Verdana;">i.e.</span></i><span style="font-family:Verdana;">, we try to recommend the best main acupuncture point for treating the disease by classifying and predicting the main acupuncture point for the disease, and further explore its acupuncture point grouping to provide the medical practitioner with the optimal solution for treating the disease and improv</span></span></span><span style="font-family:Verdana;">ing</span><span style="font-family:""><span style="font-family:Verdana;"> the clinical decision-making ability. The Bert-Chinese-Acupoint model was constructed by retraining </span><span style="font-family:Verdana;">on the basis of</span><span style="font-family:Verdana;"> the Bert model, and the semantic features in terms of acupuncture points were added to the acupunctu</span></span><span style="font-family:""><span style="font-family:Verdana;">re point corpus in the fine-tuning process to increase the semantic features in terms of acupuncture </span><span style="font-family:Verdana;">points,</span><span style="font-family:Verdana;"> and compared with the machine learning method. The results show that the Bert-Chinese Acupoint model proposed in this paper has a 3% improvement in accuracy compared to the </span><span style="font-family:Verdana;">best performing</span><span style="font-family:Verdana;"> model in the machine learning approach.
基金The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University for funding this work through Small Groups Project under Grant Number(120/43)Princess Nourah bint Abdulrahman University Researchers Supporting Project Number(PNURSP2022R281)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.The authors would like to thank the Deanship of Scientific Research at Umm Al-Qura University for supporting this work by Grant Code:(22UQU4331004DSR32).
文摘Nowadays,the usage of socialmedia platforms is rapidly increasing,and rumours or false information are also rising,especially among Arab nations.This false information is harmful to society and individuals.Blocking and detecting the spread of fake news in Arabic becomes critical.Several artificial intelligence(AI)methods,including contemporary transformer techniques,BERT,were used to detect fake news.Thus,fake news in Arabic is identified by utilizing AI approaches.This article develops a new hunterprey optimization with hybrid deep learning-based fake news detection(HPOHDL-FND)model on the Arabic corpus.The HPOHDL-FND technique undergoes extensive data pre-processing steps to transform the input data into a useful format.Besides,the HPOHDL-FND technique utilizes long-term memory with a recurrent neural network(LSTM-RNN)model for fake news detection and classification.Finally,hunter prey optimization(HPO)algorithm is exploited for optimal modification of the hyperparameters related to the LSTM-RNN model.The performance validation of the HPOHDL-FND technique is tested using two Arabic datasets.The outcomes exemplified better performance over the other existing techniques with maximum accuracy of 96.57%and 93.53%on Covid19Fakes and satirical datasets,respectively.
基金Taif University,Taif,Saudi Arabia through Taif University Researchers Supporting Project Number(TURSP-2020/115).
文摘Neural networks play a significant role in the field of image classification.When an input image is modified by adversarial attacks,the changes are imperceptible to the human eye,but it still leads to misclassification of the images.Researchers have demonstrated these attacks to make production self-driving cars misclassify StopRoad signs as 45 Miles Per Hour(MPH)road signs and a turtle being misclassified as AK47.Three primary types of defense approaches exist which can safeguard against such attacks i.e.,Gradient Masking,Robust Optimization,and Adversarial Example Detection.Very few approaches use Generative Adversarial Networks(GAN)for Defense against Adversarial Attacks.In this paper,we create a new approach to defend against adversarial attacks,dubbed Chained Dual-Generative Adversarial Network(CD-GAN)that tackles the defense against adversarial attacks by minimizing the perturbations of the adversarial image using iterative oversampling and undersampling using GANs.CD-GAN is created using two GANs,i.e.,CDGAN’s Sub-ResolutionGANandCDGAN’s Super-ResolutionGAN.The first is CDGAN’s Sub-Resolution GAN which takes the original resolution input image and oversamples it to generate a lower resolution neutralized image.The second is CDGAN’s Super-Resolution GAN which takes the output of the CDGAN’s Sub-Resolution and undersamples,it to generate the higher resolution image which removes any remaining perturbations.Chained Dual GAN is formed by chaining these two GANs together.Both of these GANs are trained independently.CDGAN’s Sub-Resolution GAN is trained using higher resolution adversarial images as inputs and lower resolution neutralized images as output image examples.Hence,this GAN downscales the image while removing adversarial attack noise.CDGAN’s Super-Resolution GAN is trained using lower resolution adversarial images as inputs and higher resolution neutralized images as output images.Because of this,it acts as an Upscaling GAN while removing the adversarial attak noise.Furthermore,CD-GAN has a modular design such that it can be prefixed to any existing classifier without any retraining or extra effort,and 2542 CMC,2023,vol.74,no.2 can defend any classifier model against adversarial attack.In this way,it is a Generalized Defense against adversarial attacks,capable of defending any classifier model against any attacks.This enables the user to directly integrate CD-GANwith an existing production deployed classifier smoothly.CD-GAN iteratively removes the adversarial noise using a multi-step approach in a modular approach.It performs comparably to the state of the arts with mean accuracy of 33.67 while using minimal compute resources in training.
基金supported by the National Natural Science Foundation of China (31871274)Natural Science Foundation of Chongqing,China (CSTB2022NSCQ-MSX0650)+2 种基金Science and Technology Research Program of Chongqing Municipal Education Commission (KJQN202100508)Team Project of Innovation Leading Talent in Chongqing (CQYC20210309536)“Contract System”Project of Chongqing Talent Plan (cstc2022ycjh-bgzxm0147)。
文摘Since the late 2010s,Artificial Intelligence(AI)including machine learning,boosted through deep learning,has boomed as a vital tool to leverage computer vision,natural language processing and speech recognition in revolutionizing zoological research.This review provides an overview of the primary tasks,core models,datasets,and applications of AI in zoological research,including animal classification,resource conservation,behavior,development,genetics and evolution,breeding and health,disease models,and paleontology.Additionally,we explore the challenges and future directions of integrating AI into this field.Based on numerous case studies,this review outlines various avenues for incorporating AI into zoological research and underscores its potential to enhance our understanding of the intricate relationships that exist within the animal kingdom.As we build a bridge between beast and byte realms,this review serves as a resource for envisioning novel AI applications in zoological research that have not yet been explored.
文摘Cardiac diseases are one of the greatest global health challenges.Due to the high annual mortality rates,cardiac diseases have attracted the attention of numerous researchers in recent years.This article proposes a hybrid fuzzy fusion classification model for cardiac arrhythmia diseases.The fusion model is utilized to optimally select the highest-ranked features generated by a variety of well-known feature-selection algorithms.An ensemble of classifiers is then applied to the fusion’s results.The proposed model classifies the arrhythmia dataset from the University of California,Irvine into normal/abnormal classes as well as 16 classes of arrhythmia.Initially,at the preprocessing steps,for the miss-valued attributes,we used the average value in the linear attributes group by the same class and the most frequent value for nominal attributes.However,in order to ensure the model optimality,we eliminated all attributes which have zero or constant values that might bias the results of utilized classifiers.The preprocessing step led to 161 out of 279 attributes(features).Thereafter,a fuzzy-based feature-selection fusion method is applied to fuse high-ranked features obtained from different heuristic feature-selection algorithms.In short,our study comprises three main blocks:(1)sensing data and preprocessing;(2)feature queuing,selection,and extraction;and(3)the predictive model.Our proposed method improves classification performance in terms of accuracy,F1measure,recall,and precision when compared to state-of-the-art techniques.It achieves 98.5%accuracy for binary class mode and 98.9%accuracy for categorized class mode.
文摘Glass is the precious material evidence of the trade of the early Silk Road. The ancient glass was easily affected by the environmental impact and weathering, and the change of composition ratios affected the correct judgment of its category. In this paper, mathematical models and methods such as Chi-square test, weighted average method, principal component analysis, cluster analysis, binary classification model and grey correlation analysis were used comprehensively to analyze the data of sample glass products combined with their categories. The results showed that the weathered high-potassium glass could be divided into 12, 9, 10 and 27, 7, 22 and so on.