Deep Learning-Based Stacked Auto-Encoder with Dynamic Differential Annealed Optimization for Skin Lesion Diagnosis

Intelligent diagnosis approaches with shallow architectural models play an essential role in healthcare.Deep Learning(DL)models with unsupervised learning concepts have been proposed because high-quality feature extraction and adequate labelled details significantly influence shallow models.On the other hand,skin lesionbased segregation and disintegration procedures play an essential role in earlier skin cancer detection.However,artefacts,an unclear boundary,poor contrast,and different lesion sizes make detection difficult.To address the issues in skin lesion diagnosis,this study creates the UDLS-DDOA model,an intelligent Unsupervised Deep Learning-based Stacked Auto-encoder(UDLS)optimized by Dynamic Differential Annealed Optimization(DDOA).Pre-processing,segregation,feature removal or separation,and disintegration are part of the proposed skin lesion diagnosis model.Pre-processing of skin lesion images occurs at the initial level for noise removal in the image using the Top hat filter and painting methodology.Following that,a Fuzzy C-Means(FCM)segregation procedure is performed using a Quasi-Oppositional Elephant Herd Optimization(QOEHO)algorithm.Besides,a novel feature extraction technique using the UDLS technique is applied where the parameter tuning takes place using DDOA.In the end,the disintegration procedure would be accomplished using a SoftMax(SM)classifier.The UDLS-DDOA model is tested against the International Skin Imaging Collaboration(ISIC)dataset,and the experimental results are examined using various computational attributes.The simulation results demonstrated that the UDLS-DDOA model outperformed the compared methods significantly.

Fault Diagnosis of Motor in Frequency Domain Signal by Stacked De-noising Auto-encoder

With the rapid development of mechanical equipment,mechanical health monitoring field has entered the era of big data.Deep learning has made a great achievement in the processing of large data of image and speech due to the powerful modeling capabilities,this also brings influence to the mechanical fault diagnosis field.Therefore,according to the characteristics of motor vibration signals(nonstationary and difficult to deal with)and mechanical‘big data’,combined with deep learning,a motor fault diagnosis method based on stacked de-noising auto-encoder is proposed.The frequency domain signals obtained by the Fourier transform are used as input to the network.This method can extract features adaptively and unsupervised,and get rid of the dependence of traditional machine learning methods on human extraction features.A supervised fine tuning of the model is then carried out by backpropagation.The Asynchronous motor in Drivetrain Dynamics Simulator system was taken as the research object,the effectiveness of the proposed method was verified by a large number of data,and research on visualization of network output,the results shown that the SDAE method is more efficient and more intelligent.

Fault Diagnosis for Rolling Bearings with Stacked Denoising Auto-encoder of Information Aggregation

Rolling bearings are important central components in rotating machines, whose fault diagnosis is crucial in condition-based maintenance to reduce the complexity of different kinds of faults. To classify various rolling bearing faults, a prognostic algorithm consisting of four phases was proposed. Since stacked denoising auto-encoder can be filtered, noise of large numbers of mechanical vibration signals was used for deep learning structure to extract the characteristics of the noise. Unsupervised pre-training method, which can greatly simplify the traditional manual extraction approach, was utilized to process the depth of the data automatically. Furthermore, the aggregation layer of stacked denoising auto-encoder(SDA) was proposed to get rid of gradient disappearance in deeper layers of network, mix superficial nodes’ expression with deeper layers, and avoid the insufficient express ability in deeper layers. Principal component analysis(PCA) was adopted to extract different features for classification. According to the experimental data of this method and from the comparison results, the proposed method of rolling bearing fault classification reached 97.02% of correct rate, suggesting a better performance than other algorithms.

Predicting the Antigenic Variant of Human Influenza A(H3N2) Virus with a Stacked Auto-Encoder Model

The influenza virus changes its antigenicity frequently due to rapid mutations, leading to immune escape and failure of vaccination. Rapid determination of the influenza antigenicity could help identify the antigenic variants in time. Here, we built a stacked auto-encoder (SAE) model for predicting the antigenic variant of human influenza A(H3N2) viruses based on the hemagglutinin (HA) protein sequences. The model achieved an accuracy of 0.95 in five-fold cross-validations, better than the logistic regression model did. Further analysis of the model shows that most of the active nodes in the hidden layer reflected the combined contribution of multiple residues to antigenic variation. Besides, some features (residues on HA protein) in the input layer were observed to take part in multiple active nodes, such as residue 189, 145 and 156, which were also reported to mostly determine the antigenic variation of influenza A(H3N2) viruses. Overall,this work is not only useful for rapidly identifying antigenic variants in influenza prevention, but also an interesting attempt in inferring the mechanisms of biological process through analysis of SAE model, which may give some insights into interpretation of the deep learning

Software Defect Prediction Based on Stacked Contractive Autoencoder and Multi-Objective Optimization

Software defect prediction plays an important role in software quality assurance.However,the performance of the prediction model is susceptible to the irrelevant and redundant features.In addition,previous studies mostly regard software defect prediction as a single objective optimization problem,and multi-objective software defect prediction has not been thoroughly investigated.For the above two reasons,we propose the following solutions in this paper:(1)we leverage an advanced deep neural network-Stacked Contractive AutoEncoder(SCAE)to extract the robust deep semantic features from the original defect features,which has stronger discrimination capacity for different classes(defective or non-defective).(2)we propose a novel multi-objective defect prediction model named SMONGE that utilizes the Multi-Objective NSGAII algorithm to optimize the advanced neural network-Extreme learning machine(ELM)based on state-of-the-art Pareto optimal solutions according to the features extracted by SCAE.We mainly consider two objectives.One objective is to maximize the performance of ELM,which refers to the benefit of the SMONGE model.Another objective is to minimize the output weight norm of ELM,which is related to the cost of the SMONGE model.We compare the SCAE with six state-of-the-art feature extraction methods and compare the SMONGE model with multiple baseline models that contain four classic defect predictors and the MONGE model without SCAE across 20 open source software projects.The experimental results verify that the superiority of SCAE and SMONGE on seven evaluation metrics.

A Feature Learning-Based Model for Analyzing Students’ Performance in Supportive Learning

Supportive learning plays a substantial role in providing a quality edu-cation system.The evaluation of students’performance impacts their deeper insight into the subject knowledge.Specifically,it is essential to maintain the baseline foundation for building a broader understanding of their careers.This research concentrates on establishing the students’knowledge relationship even in reduced samples.Here,Synthetic Minority Oversampling TEchnique(SMOTE)technique is used for pre-processing the missing value in the provided input dataset to enhance the prediction accuracy.When the initial processing is not done substantially,it leads to misleading prediction accuracy.This research concentrates on modelling an efficient classifier model to predict students’perfor-mance.Generally,the online available student dataset comprises a lesser amount of sample,and k-fold cross-validation is performed to balance the dataset.Then,the relationship among the students’performance(features)is measured using the auto-encoder.The stacked Long Short Term Memory(s-LSTM)is used to learn the previous feedback connection.The stacked model handles the provided data and the data sequence for understanding the long-term dependencies.The simula-tion is done in the MATLAB 2020a environment,and the proposed model shows a better trade-off than the existing approaches.Some evaluation metrics like pre-diction accuracy,sensitivity,specificity,AUROC,F1-score and recall are evalu-ated using the proposed model.The performance of the s?LSTM model is compared with existing approaches.The proposed model gives 89% accuracy,83% precision,86%recall,and 87%F-score.The proposed model outperforms the existing systems in terms of the earlier metrics.

Early Detection of Heartbeat from Multimodal Data Using RPA Learning with KDNN-SAE

Heartbeat detection stays central to cardiovascular an electrocardiogram(ECG)is used to help with disease diagnosis and management.Existing Convolutional Neural Network(CNN)-based methods suffer from the less generalization problem thus;the effectiveness and robustness of the traditional heartbeat detector methods cannot be guaranteed.In contrast,this work proposes a heartbeat detector Krill based Deep Neural Network Stacked Auto Encoders(KDNN-SAE)that computes the disease before the exact heart rate by combining features from multiple ECG Signals.Heartbeats are classified independently and multiple signals are fused to estimate life threatening conditions earlier without any error in classification of heart beat.This work contained Training and testing stages,in the preparation part at first the Adaptive Filter Enthalpy-based Empirical Mode Decomposition(EMD)is utilized to eliminate the motion artifact in the signal.At that point,the robotic process automation(RPA)learning part extracts the effective features are extracted,and normalized the value of the feature then estimated utilizing the RPA loss function.At last KDNN-SAE prepared training for the data stored in the dataset.In the subsequent stage,input signal compute motion artifact and RPA Learning the evaluation part determines the detection of Heartbeat.So early diagnosis of heart failures is an essential factor.The results of the experiments show that our proposed method has a high score outcome of 0.9997.Comparable to the CIF,which reaches 0.9990.The CNN and Artificial Neural Network(ANN)had less score 0.95115 and 0.90147.

混合深层协同过滤的SVD++推荐方法

为抑制辅助信息在推荐模型中各个方向的扰动并考虑使用文本信息提取项目特征,提出一种矩阵分解模型,混合深层协同过滤的SVD++推荐方法.该模型将附加栈式降噪自编码器和堆叠的收缩降噪自编码器与辅助信息相结合,分别提取用户和项目的潜在特征表示,并在提取项目特征表示时加入预训练的词嵌入模型考虑词语之间的语义关系.在数据集MovieLens-1M与MovieLens-10M的实验.结果表明:相比于传统算法、深度学习算法以及所提模型的变体,所提模型更有效地提取潜在特征表示并提高预测评分精度.

冲击加载条件下AZ31B镁合金变形孪晶的微观特征

采用分离式Hopkinson压杆(SHPB)对AZ31B热轧板材沿轧向进行冲击实验,加载应变率约为1 200 s-1.利用金相显微镜和透射电子显微镜观察微观组织特征。结果表明:在应变量约为0.003的条件下,压缩孪生{101-1}/<101-2>与拉伸孪生{11-02}/<1 1-01>均为孪生塑性变形机制,这与准静态加载下单一的拉伸孪生机制不同。孪晶内部的微观特征演化过程包括:1)平行排列基面层错的出现;2)与孪晶界成特定角度平行排列位错线的形成;3)位错胞的产生。

基于数据均衡的增进式深度自动图像标注

自动图像标注是一个包含众多标签、多样特征的富有挑战性的研究问题,是新一代图像检索与图像理解的关键步骤.针对传统的基于浅层机器学习标注算法标注效率低下、难以处理复杂分类任务的问题,提出了基于栈式自动编码器(stacked auto-encoder,简称SAE)的自动图像标注算法,提升了标注效率和标注效果.主要针对图像标注数据不平衡问题,提出两种解决思路:对于标注模型,提出一种增强训练中低频标签的平衡栈式自动编码器(B-SAE),较好地改善了中低频标签的标注效果.并在该模型的基础上提出一种分组强化训练B-SAE子模型的鲁棒平衡栈式自动编码器算法(RB-SAE),提升了标注的稳定性,从而保证模型本身具有较强的处理不平衡数据的能力;对于标注过程,以未知图像作为出发点,首先构造未知图像的局部均衡数据集,并判定该图像的高低频属性以决定不同的标注过程,局部语义传播算法(SP)标注中低频图像,RB-SAE算法标注高频图像,形成属性判别的标注框架(ADA),保证了标注过程具有较强的应对不平衡数据的能力,从而提升整体图像标注效果.通过在3个公共数据集上进行实验验证,结果表明,该方法在许多指标上相比以往方法均有较大提高.

基于堆栈压缩自编码的近红外光谱药品鉴别方法

由于近红外光谱在药品鉴别应用中具有分析速度快、样品无损、可现场检测等突出优点,目前已在众多领域中广泛应用。但近红外光谱存在信噪比低,吸收强度弱且谱峰重叠等缺点,无法从光谱中直接得到定性/定量的物质信息,因而近红外光谱分析技术常作为一种间接分析技术,并且光谱的化学计量学建模方法成为近红外光谱分析的核心内容。深度学习是机器学习的一个新的分支,并已经成功运用于多个领域。深度学习的网络结构和非线性的激活能力,使其模型特别适合高维、非线性的大规模数据建模。为进一步丰富近红外光谱建模方法,并提高近红外光谱分析技术的回归精度或分类准确率,将深度学习方法应用于近红外光谱分析,发展新的建模方法十分必要。面向近红外光谱定性分析技术,提出一种基于堆栈压缩自编码网络(SCAE)光谱定性分析方法,并应用于多类别药品的光谱分析,以区分或鉴别不同厂家生产的同种药品。压缩自编码网络(CAE)以自编码网络(AE)为基础,进一步加入雅克比矩阵作为约束项。自编码网络最初是用实现数据降维,以学习数据内部特征,而雅克比矩阵包含数据在各个方向上的信息,将其作为AE的约束项则可使提取到的特征对输入数据在一定程度下的扰动具有不变性,从而提高AE提取特征的能力。SCAE是一种由多层CAE构成的神经网络。前一层CAE的隐藏层作为后一层CAE的输入层,网络的全部参数是通过采用逐层贪婪的训练方式来获取的,训练结束后将所有网络视为一个整体,通过反向传播算法进行微调,最后使用Logistic/Softmax分类器进行定性分析。实验数据均为中国食品药品检定研究院采集,以头孢克肟胶囊作为二分类实验数据,硝酸异山梨酯片作为多分类实验数据。通过Bruker Matrix光谱仪测定每个样本在不同波长下的吸光度值得到其光谱曲线,再通过OPUS软件消除漂移等因素对光谱样本之间产生的偏差。接下来通过实验确定约束项雅克比矩阵的系数λ为0.003之后建立模型。建模过程分为五个阶段,分别为:预处理阶段,预训练阶段,微调阶段,测试阶段和对比阶段。为了验证SCAE在分类准确性、算法稳定性和建模时间等方面的性能,与BP神经网络、SVM算法、稀疏自编码(SAE)和降噪自编码(DAE)开展对比实验研究。分类准确性方面,在不同的训练集与测试集的比例下,SCAE均有最佳的分类准确性与算法稳定性。建模时间方面,由于SVM算法不需要预训练和特征提取,所以运行时间方面比其他算法有大的优势,但是SCAE建模速度优于除SVM之外的其他对比算法。综合而言,使用SCAE进行药品鉴别有效可行。

基于环氧树脂灌封的三维叠层组件裂纹问题分析与对策研究

对基于环氧树脂灌封的三维叠层组件经过温度循环等考核后出现裂纹的机理进行探讨,通过分析认为,裂纹是灌封过程中引入的固有缺陷,后续在固化过程中收缩应力及热应力的作用下形成.并从优化固化条件、提高材料间的热匹配性以及改善灌封工艺等方面提出具体的解决措施,提高了三维叠层组件的可靠性.

基于SCAE-ACGAN的直升机行星齿轮裂纹故障诊断

直升机行星传动轮系结构复杂、工况多变,其振动信号受工况影响大,在故障样本较少的情况下导致行星齿轮箱故障诊断准确率不高,早期故障诊断困难。针对上述问题,提出将堆栈收缩自动编码网络(stacked contractive autoencoder,简称SCAE)与辅助分类生成式对抗网络(auxiliary classifier generative adversarial networks,简称ACGAN)相结合的SCAE-ACGAN故障诊断方法。ACGAN的生成器产生与真实样本具有类似分布的生成样本,扩展训练样本集,并与真实样本一起输入至判别器进行训练。ACGAN采用SCAE作为判别器,利用SCAE良好的抗数据波动能力,从扩展样本集中挖掘出有效的深度特征,并实现样本的真伪与类别的判定。ACGAN的判别器和生成器在对抗学习训练机制下交替优化,提高方法的样本生成质量与故障判定能力。将SCAE-ACGAN应用于直升机行星轮裂纹故障诊断,结果表明,SCAE-ACGAN的故障诊断性能好,在样本数量少与工况变化情况下具有较好的健壮性和适应性。

A New Method for Sentiment Analysis Using Contextual Auto-Encoders

Sentiment analysis, a hot research topic, presents new challenges for understanding users＇ opinions and judg-ments expressed online. They aim to classify the subjective texts by assigning them a polarity label. In this paper, weintroduce a novel machine learning framework using auto-encoders network to predict the sentiment polarity label at theword level and the sentence level. Inspired by the dimensionality reduction and the feature extraction capabilities of theauto-encoders, we propose a new model for distributed word vector representation ＂PMI-SA＂ using as input pointwise-mutual-information ＂PMI＂ word vectors. The resulted continuous word vectors are combined to represent a sentence. Anunsupervised sentence embedding method, called Contextual Recursive Auto-Encoders ＂CoRAE＂, is also developed forlearning sentence representation. Indeed, CoRAE follows the basic idea of the recursive auto-encoders to deeply composethe vectors of words constituting the sentence, but without relying on any syntactic parse tree. The CoRAE model consistsin combining recursively each word with its context words （neighbors＇ words： previous and next） by considering the wordorder. A support vector machine classifier with fine-tuning technique is also used to show that our deep compositionalrepresentation model CoRAE improves significantly the accuracy of sentiment analysis task. Experimental results demon-strate that CoRAE remarkably outperforms several competitive baseline methods on two databases, namely, Sanders twittercorpus and Facebook comments corpus. The CoRAE model achieves an efficiency of 83.28% with the Facebook dataset and97.57% with the Sanders dataset.

基于改进的TFIDF和压缩自动编码器文本分类研究

为了提高文本分类的分类效果和降低分类的错误率,本文将深度学习中的压缩自动编码器逐层叠加,提出基于改进的TFIDF和堆叠的压缩自动编码器SCAE(Stack Contractive Auto-Encoder)的文本分类思想,将SCAE构成深度神经网络,无监督的训练学习文本,提高特征提取的鲁棒性,并使用反向传播算法优化网络中的参数,在计算特征词的权重时,采用本文改进的TFIDF方法。通过实验将CAE和SAE(稀疏自动编码器)进行比较,采用支持向量机(SVM)分类。实验表明,单层的CAE比单层的SAE的分类性能更好,堆叠压缩编码器学习比堆叠的稀疏编码器的分类性能同样要好。

一种基于栈式压缩自编码的高光谱图像分类方法

针对高光谱图像传统分类方法精度低、模型稳定性差而深度学习模型时间消耗大、计算成本高的问题,充分考虑高光谱图像的光谱信息和空间信息,提出了一种基于栈式压缩自编码的高光谱图像分类方法。将提取的邻域空间信息与待分类像素点的光谱信息融合,利用栈式压缩自编码提取融合后信息的深层特征,再利用逻辑回归确定高光谱图像中各像素点的类别。该方法在Indian Pines和Pavia University数据集上的总体分类精度分别达到了89.943%、93.949%。相比其他方法,该方法分类性能更优,可用于高光谱图像分类。

Deep learning model for estimating the mechanical properties of concrete containing silica fume exposed to high temperatures

In this study,the deep learning models for estimating the mechanical properties of concrete containing silica fume subjected to high temperatures were devised.Silica fume was used at concentrations of 0%,5%,10%,and 20%.Cube specimens(100 mm×100 mm×100 mm)were prepared for testing the compressive strength and ultrasonic pulse velocity.They were cured at 20℃zb2℃ in a standard cure for 7,28,and 90 d.After curing,they were subjected to temperatures of 20℃,200℃,400℃,600℃,and 800℃.Two well-known deep learning approaches,i.e.,stacked autoencoders and long short-term memory(LSTM)networks,were used for forecasting the compressive strength and ultrasonic pulse velocity of concrete containing silica fume subjected to high temperatures.The forecasting experiments were carried out using MATLAB deep learning and neural network tools,respectively.Various statistical measures were used to validate the prediction performances of both the approaches.This study found that the LSTM network achieved better results than the stacked autoencoders.In addition,this study found that deep learning,which has a very good prediction ability with little experimental data,was a convenient method for civil engineering.

锌垛打捆质量的改善

打捆是锌冶炼行业熔铸工段中最重要的工序,打捆质量的好坏关系到了整个锌垛质量的优劣,影响了锌产品的运输和销售。为此对几个影响锌垛打捆质量的原因进行分析,并提出了相应的解决方案。