Stacking Ensemble Learning-Based Convolutional Gated Recurrent Neural Network for Diabetes Miletus

Diabetes mellitus is a metabolic disease in which blood glucose levels rise as a result of pancreatic insulin production failure.It causes hyperglycemia and chronic multiorgan dysfunction,including blindness,renal failure,and cardi-ovascular disease,if left untreated.One of the essential checks that are needed to be performed frequently in Type 1 Diabetes Mellitus is a blood test,this procedure involves extracting blood quite frequently,which leads to subject discomfort increasing the possibility of infection when the procedure is often recurring.Exist-ing methods used for diabetes classification have less classification accuracy and suffer from vanishing gradient problems,to overcome these issues,we proposed stacking ensemble learning-based convolutional gated recurrent neural network(CGRNN)Metamodel algorithm.Our proposed method initially performs outlier detection to remove outlier data,using the Gaussian distribution method,and the Box-cox method is used to correctly order the dataset.After the outliers’detec-tion,the missing values are replaced by the data’s mean rather than their elimina-tion.In the stacking ensemble base model,multiple machine learning algorithms like Naïve Bayes,Bagging with random forest,and Adaboost Decision tree have been employed.CGRNN Meta model uses two hidden layers Long-Short-Time Memory(LSTM)and Gated Recurrent Unit(GRU)to calculate the weight matrix for diabetes prediction.Finally,the calculated weight matrix is passed to the soft-max function in the output layer to produce the diabetes prediction results.By using LSTM-based CG-RNN,the mean square error(MSE)value is 0.016 and the obtained accuracy is 91.33%.

Mapping Network-Coordinated Stacked Gated Recurrent Units for Turbulence Prediction

Accurately predicting fluid forces acting on the sur-face of a structure is crucial in engineering design.However,this task becomes particularly challenging in turbulent flow,due to the complex and irregular changes in the flow field.In this study,we propose a novel deep learning method,named mapping net-work-coordinated stacked gated recurrent units(MSU),for pre-dicting pressure on a circular cylinder from velocity data.Specifi-cally,our coordinated learning strategy is designed to extract the most critical velocity point for prediction,a process that has not been explored before.In our experiments,MSU extracts one point from a velocity field containing 121 points and utilizes this point to accurately predict 100 pressure points on the cylinder.This method significantly reduces the workload of data measure-ment in practical engineering applications.Our experimental results demonstrate that MSU predictions are highly similar to the real turbulent data in both spatio-temporal and individual aspects.Furthermore,the comparison results show that MSU predicts more precise results,even outperforming models that use all velocity field points.Compared with state-of-the-art methods,MSU has an average improvement of more than 45%in various indicators such as root mean square error(RMSE).Through comprehensive and authoritative physical verification,we estab-lished that MSU’s prediction results closely align with pressure field data obtained in real turbulence fields.This confirmation underscores the considerable potential of MSU for practical applications in real engineering scenarios.The code is available at https://github.com/zhangzm0128/MSU.

Fake News Classification Using a Fuzzy Convolutional Recurrent Neural Network

In recent years,social media platforms have gained immense popularity.As a result,there has been a tremendous increase in content on social media platforms.This content can be related to an individual’s sentiments,thoughts,stories,advertisements,and news,among many other content types.With the recent increase in online content,the importance of identifying fake and real news has increased.Although,there is a lot of work present to detect fake news,a study on Fuzzy CRNN was not explored into this direction.In this work,a system is designed to classify fake and real news using fuzzy logic.The initial feature extraction process is done using a convolutional recurrent neural network(CRNN).After the extraction of features,word indexing is done with high dimensionality.Then,based on the indexing measures,the ranking process identifies whether news is fake or real.The fuzzy CRNN model is trained to yield outstanding resultswith 99.99±0.01%accuracy.This work utilizes three different datasets(LIAR,LIAR-PLUS,and ISOT)to find the most accurate model.

Recurrent Convolutional Neural Network MSER-Based Approach for Payable Document Processing

A tremendous amount of vendor invoices is generated in the corporate sector.To automate the manual data entry in payable documents,highly accurate Optical Character Recognition(OCR)is required.This paper proposes an end-to-end OCR system that does both localization and recognition and serves as a single unit to automate payable document processing such as cheques and cash disbursement.For text localization,the maximally stable extremal region is used,which extracts a word or digit chunk from an invoice.This chunk is later passed to the deep learning model,which performs text recognition.The deep learning model utilizes both convolution neural networks and long short-term memory(LSTM).The convolution layer is used for extracting features,which are fed to the LSTM.The model integrates feature extraction,modeling sequence,and transcription into a unified network.It handles the sequences of unconstrained lengths,independent of the character segmentation or horizontal scale normalization.Furthermore,it applies to both the lexicon-free and lexicon-based text recognition,and finally,it produces a comparatively smaller model,which can be implemented in practical applications.The overall superior performance in the experimental evaluation demonstrates the usefulness of the proposed model.The model is thus generic and can be used for other similar recognition scenarios.

Lightweight and highly robust memristor-based hybrid neural networks for electroencephalogram signal processing

Memristor-based neuromorphic computing shows great potential for high-speed and high-throughput signal processing applications,such as electroencephalogram(EEG)signal processing.Nonetheless,the size of one-transistor one-resistor(1T1R)memristor arrays is limited by the non-ideality of the devices,which prevents the hardware implementation of large and complex networks.In this work,we propose the depthwise separable convolution and bidirectional gate recurrent unit(DSC-BiGRU)network,a lightweight and highly robust hybrid neural network based on 1T1R arrays that enables efficient processing of EEG signals in the temporal,frequency and spatial domains by hybridizing DSC and BiGRU blocks.The network size is reduced and the network robustness is improved while ensuring the network classification accuracy.In the simulation,the measured non-idealities of the 1T1R array are brought into the network through statistical analysis.Compared with traditional convolutional networks,the network parameters are reduced by 95%and the network classification accuracy is improved by 21%at a 95%array yield rate and 5%tolerable error.This work demonstrates that lightweight and highly robust networks based on memristor arrays hold great promise for applications that rely on low consumption and high efficiency.

Multi-Scale Convolutional Gated Recurrent Unit Networks for Tool Wear Prediction in Smart Manufacturing

As an integrated application of modern information technologies and artificial intelligence,Prognostic and Health Management(PHM)is important for machine health monitoring.Prediction of tool wear is one of the symbolic applications of PHM technology in modern manufacturing systems and industry.In this paper,a multi-scale Convolutional Gated Recurrent Unit network(MCGRU)is proposed to address raw sensory data for tool wear prediction.At the bottom of MCGRU,six parallel and independent branches with different kernel sizes are designed to form a multi-scale convolutional neural network,which augments the adaptability to features of different time scales.These features of different scales extracted from raw data are then fed into a Deep Gated Recurrent Unit network to capture long-term dependencies and learn significant representations.At the top of the MCGRU,a fully connected layer and a regression layer are built for cutting tool wear prediction.Two case studies are performed to verify the capability and effectiveness of the proposed MCGRU network and results show that MCGRU outperforms several state-of-the-art baseline models.

Improved Quantile Convolutional and Recurrent Neural Networks for Electric Vehicle Battery Temperature Prediction

The battery thermal management of electric vehicles can be improved using neural networks predicting quantile sequences of the battery temperature.This work extends a method for the development of Quantile Convolutional and Quantile Recurrent Neural Networks(namely Q*NN).Fleet data of 225629 drives are clustered and balanced,simulation data from 971 simulations are augmented before they are combined for training and testing.The Q*NN hyperparameters are optimized using an efficient Bayesian optimization,before the Q*NN models are compared with regression and quantile regression models for four horizons.The analysis of point-forecast and quantile-related metrics shows the superior performance of the novel Q*NN models.The median predictions of the best performing model achieve an average RMSE of 0.66°C and R^(2) of 0.84.The predicted 0.99 quantile covers 98.87%of the true values in the test data.In conclusion,this work proposes an extended development and comparison of Q*NN models for accurate battery temperature prediction.

Enhancing Human Action Recognition with Adaptive Hybrid Deep Attentive Networks and Archerfish Optimization

In recent years,wearable devices-based Human Activity Recognition(HAR)models have received significant attention.Previously developed HAR models use hand-crafted features to recognize human activities,leading to the extraction of basic features.The images captured by wearable sensors contain advanced features,allowing them to be analyzed by deep learning algorithms to enhance the detection and recognition of human actions.Poor lighting and limited sensor capabilities can impact data quality,making the recognition of human actions a challenging task.The unimodal-based HAR approaches are not suitable in a real-time environment.Therefore,an updated HAR model is developed using multiple types of data and an advanced deep-learning approach.Firstly,the required signals and sensor data are accumulated from the standard databases.From these signals,the wave features are retrieved.Then the extracted wave features and sensor data are given as the input to recognize the human activity.An Adaptive Hybrid Deep Attentive Network(AHDAN)is developed by incorporating a“1D Convolutional Neural Network(1DCNN)”with a“Gated Recurrent Unit(GRU)”for the human activity recognition process.Additionally,the Enhanced Archerfish Hunting Optimizer(EAHO)is suggested to fine-tune the network parameters for enhancing the recognition process.An experimental evaluation is performed on various deep learning networks and heuristic algorithms to confirm the effectiveness of the proposed HAR model.The EAHO-based HAR model outperforms traditional deep learning networks with an accuracy of 95.36,95.25 for recall,95.48 for specificity,and 95.47 for precision,respectively.The result proved that the developed model is effective in recognizing human action by taking less time.Additionally,it reduces the computation complexity and overfitting issue through using an optimization approach.

Deep Convolutional Neural Network Based Churn Prediction for Telecommunication Industry

Currently,mobile communication is one of the widely used means of communication.Nevertheless,it is quite challenging for a telecommunication company to attract new customers.The recent concept of mobile number portability has also aggravated the problem of customer churn.Companies need to identify beforehand the customers,who could potentially churn out to the competitors.In the telecommunication industry,such identification could be done based on call detail records.This research presents an extensive experimental study based on various deep learning models,such as the 1D convolutional neural network(CNN)model along with the recurrent neural network(RNN)and deep neural network(DNN)for churn prediction.We use the mobile telephony churn prediction dataset obtained from customers-dna.com,containing the data for around 100,000 individuals,out of which 86,000 are non-churners,whereas 14,000 are churned customers.The imbalanced data are handled using undersampling and oversampling.The accuracy for CNN,RNN,and DNN is 91%,93%,and 96%,respectively.Furthermore,DNN got 99%for ROC.

基于ASTG-CRNN模型的多步长交通流预测

针对交通流预测模型中路网表征结构难以进行刻画和交通流数据中动态时空相关性难以进行建模以及其中时间特征捕获不充分的问题,提出一种基于注意力机制和时空图卷积循环神经网络的交通流预测模型(ASTG-CRNN)。首先,通过定义节点相对邻近度来确定路网表征结构的关系权重;其次,通过在时空维度上引入注意力机制对动态时空相关性进行建模,再采用图卷积捕获交通流数据中的空间特征;最后,采用卷积神经网络和双向门控循环神经单元的组合模块共同捕捉时间特征,从而能更好地表达交通流的时空特性。在两个公开交通流数据集PeMS04和PeMS08上对模型预测效果进行验证,其结果表明,ASTG-CRNN模型的预测结果均优于其它模型,与时空同步图卷积网络模型(STSGCN)相比,在未来1h内预测结果的MAE、RMSE和MAPE在数据集PeMS08上分别降低了2.71、2.69和0.87%。

Deep learning neural networks for spatially explicit prediction of flash flood probability

Flood probability maps are essential for a range of applications,including land use planning and developing mitigation strategies and early warning systems.This study describes the potential application of two architectures of deep learning neural networks,namely convolutional neural networks(CNN)and recurrent neural networks(RNN),for spatially explicit prediction and mapping of flash flood probability.To develop and validate the predictive models,a geospatial database that contained records for the historical flood events and geo-environmental characteristics of the Golestan Province in northern Iran was constructed.The step-wise weight assessment ratio analysis(SWARA)was employed to investigate the spatial interplay between floods and different influencing factors.The CNN and RNN models were trained using the SWARA weights and validated using the receiver operating characteristics technique.The results showed that the CNN model(AUC=0.832,RMSE=0.144)performed slightly better than the RNN model(AUC=0.814,RMSE=0.181)in predicting future floods.Further,these models demonstrated an improved prediction of floods compared to previous studies that used different models in the same study area.This study showed that the spatially explicit deep learning neural network models are successful in capturing the heterogeneity of spatial patterns of flood probability in the Golestan Province,and the resulting probability maps can be used for the development of mitigation plans in response to the future floods.The general policy implication of our study suggests that design,implementation,and verification of flood early warning systems should be directed to approximately 40%of the land area characterized by high and very susceptibility to flooding.

基于双图转换和融合CRNN网络的轴承故障诊断

针对一维振动序列输入制约卷积神经网络性能,且单一数据处理方法限制实际复杂工况下轴承故障特性的深层挖掘等问题,提出了一种基于双图转换与多卷积循环神经网络融合的滚动轴承故障诊断方法。分别利用格拉姆角差场和马尔可夫转移场编码方法将一维序列信号转换为二维图像。将转换后的两种模态图像同时输入多CRNN融合的Fu-CRNN网络模型,充分汲取两种转换方法优点并提高CRNN模型特征表达能力。实现轴承信号特征自适应提取及端到端诊断。为验证该方法的可靠性与优越性,选用凯斯西储大学滚动轴承数据集进行轴承故障诊断试验,并比较分析诊断性能。结果表明,所提模型识别准确率和泛化效果均优于单一模态样本输入模型,相较于其他常用算法表现更出色,可为样本构建和轴承故障诊断方法提供参考。

Multimodal emotion recognition based on deep neural network

In order to increase the accuracy rate of emotion recognition in voiceand video,the mixed convolutional neural network(CNN)and recurrent neural network(RNN)ae used to encode and integrate the two information sources.For the audio signals,several frequency bands as well as some energy functions are extacted as low-level features by using a sophisticated audio technique,and then they are encoded w it a one-dimensional(I D)convolutional neural network to abstact high-level features.Finally,tiese are fed into a recurrent neural network for te sake of capturing dynamic tone changes in a temporal dimensionality.As a contrast,a two-dimensional(2D)convolutional neural network and a similar RNN are used to capture dynamic facial appearance changes of temporal sequences.The method was used in te Chinese Natral Audio-'Visual Emotion Database in te Chinese Conference on Pattern Recognition(CCPR)in2016.Experimental results demonstrate that te classification average precision of the proposed metiod is41.15%,which is increased by16.62%compaed with te baseline algorithm offered by the CCPR in2016.It is proved ta t te proposed method has higher accuracy in te identification of emotional information.

基于DP-DBNet和MHA-CRNN的船牌号检测与识别

船牌号的检测和识别对于港口的智能化管理和解决传统人工方式监管渔船中存在的耗时耗力的问题具有重要意义.针对船牌悬挂位置,背景颜色和字符个数不统一等特点,本文提出两阶段双模型的检测和识别方法.首先,提出将双路径网络(dual path networks, DPN)与可微二值化网络(differentiable binarization network, DBNet)相结合的DP-DBNet船牌号位置检测模型.其次,提出将多头注意力机制(multi-head-attention mechanism, MHA)与改进的卷积循环神经网络(convolutional recurrent neural network, CRNN)相结合的MHA-CRNN船牌号文字识别模型.最后,以烟台芝罘区新型现代化智慧渔港项目为数据来源,并进行算法对比实验分析;实验结果表明,两种模型结合的两阶段识别方法可以使船牌号的识别准确率达到76.39%,充分证明了该模型的有效性和在海洋港口管理方面的应用价值.

改进DBNet与CRNN的面标识别方法

为解决板坯喷涂面标实时识别问题,构建文本检测和识别模型。改进可微二值化算法网络,引入高效通道注意力模块SENet,进行自适应空间特征融合(ASFF),增强特征金字塔预测多尺度目标的能力。识别模型改进卷积递归神经网络的VGG网络,将卷积与循环神经网络联合训练。实验结果表明,检测模型的精确率、召回率和调和平均值达到93.30%、86.45%、89.85%,提升显著;识别模型平均准确率达到86.01%,精度提升4.99%。模型满足实时与准确性要求。

Deep Neural Network-Based Chinese Semantic Role Labeling

A recent trend in machine learning is to use deep architectures to discover multiple levels of features from data,which has achieved impressive results on various natural language processing(NLP)tasks.We propose a deep neural network-based solution to Chinese semantic role labeling(SRL)with its application on message analysis.The solution adopts a six-step strategy:text normalization,named entity recognition(NER),Chinese word segmentation and part-of-speech(POS)tagging,theme classification,SRL,and slot filling.For each step,a novel deep neural network-based model is designed and optimized,particularly for smart phone applications.Experiment results on all the NLP sub-tasks of the solution show that the proposed neural networks achieve state-of-the-art performance with the minimal computational cost.The speed advantage of deep neural networks makes them more competitive for large-scale applications or applications requiring real-time response,highlighting the potential of the proposed solution for practical NLP systems.

融合CTPN和CRNN对识别影像图片文字及应用的研究

对生物医学领域中影像图片识别的问题进行了研究,提出了融合CTPN和CRNN模型的影像图片文字识别方法。首先利用CTPN模型检测文字所在位置,再利用卷积循环神经网络进行特征序列提取和识别,并基于BLSTM学习序列特征将得分高的窗口连接作为文本区域,最后得到文本的检测识别结果。研究结果表明,融合后能够高效地定位和识别医学影像中的文字信息,为医学影像图片的处理提供了一种有效的解决方案。

Text Understanding with a Hybrid Neural Network Based Learning

Teaching machine to understand needs to design an algorithm for the machine to comprehend documents. As some traditional methods cannot learn the inherent characters effectively, this paper presents a new hybrid neural network model to extract sentence-level summarization from single document,and it allows us to develop an attention based deep neural network that can learn to understand documents with minimal prior knowledge. The proposed model composed of multiple processing layers can learn the representations of features.Word embedding is used to learn continuous word representations for constructing sentence as input to convolutional neural network. The recurrent neural network is also used to label the sentences from the original document, and the proposed BAM-GRU model is more efficient. Experimental results show the feasibility of the approach. Some problems and further works are also present in the end.

Deep-fake video detection approaches using convolutional–recurrent neural networks

Deep-Fake is an emerging technology used in synthetic media which manipulates individuals in existing images and videos with someone else’s likeness.This paper presents the comparative study of different deep neural networks employed for Deep-Fake video detection.In the model,the features from the training data are extracted with the intended Convolution Neural Network model to form feature vectors which are further analysed using a dense layer,a Long Short-Term Memoryand Gated Recurrent by adopting transfer learning with fine tuning for training the models.The model is evaluated to detect Artificial Intelligence based Deep fakes images and videos using benchmark datasets.Comparative analysis shows that the detections are majorly biased towards domain of the dataset but there is a noteworthy improvement in the model performance parameters by using Transfer Learning whereas Convolutional-Recurrent Neural Network has benefits in sequence detection.

融合CNN-BiGRU和注意力机制的网络入侵检测模型

为提高网络入侵检测模型特征提取能力和分类准确率,提出了一种融合双向门控循环单元(CNN-BiGRU)和注意力机制的网络入侵检测模型.使用CNN有效提取流量数据集中的非线性特征;双向门控循环单元(BiGRU)提取数据集中的时序特征,最后融合注意力机制对不同类型流量数据通过加权的方式进行重要程度的区分,从而整体提高该模型特征提取与分类的性能.实验结果表明:其整体精确率比双向长短期记忆网络(BiLSTM)模型提升了2.25%.K折交叉验证结果表明:该模型泛化性能良好,避免了过拟合现象的发生,印证了该模型的有效性与合理性.