A RECURSIVE ALGORITHM AND ITS CONVERGENCE FOR PARAMETER ESTIMATION OF CONVOLUTION MODEL

In this article, the problem on the estimation of the convolution model parameters is considered. The recursive algorithm for estimating model parameters is introduced from the orthogonal procedure of the data, the convergence of this algorithm is theoretically discussed, and a sufficient condition for the convergence criterion of the orthogonal procedure is given. According to this condition, the recursive algorithm is convergent to model wavelet A- = （1, α1,..., αq）.

User Churn Prediction Hierarchical Model Based on Graph Attention Convolutional Neural Networks

The telecommunications industry is becoming increasingly aware of potential subscriber churn as a result of the growing popularity of smartphones in the mobile Internet era,the quick development of telecommunications services,the implementation of the number portability policy,and the intensifying competition among operators.At the same time,users'consumption preferences and choices are evolving.Excellent churn prediction models must be created in order to accurately predict the churn tendency,since keeping existing customers is far less expensive than acquiring new ones.But conventional or learning-based algorithms can only go so far into a single subscriber's data;they cannot take into consideration changes in a subscriber's subscription and ignore the coupling and correlation between various features.Additionally,the current churn prediction models have a high computational burden,a fuzzy weight distribution,and significant resource economic costs.The prediction algorithms involving network models currently in use primarily take into account the private information shared between users with text and pictures,ignoring the reference value supplied by other users with the same package.This work suggests a user churn prediction model based on Graph Attention Convolutional Neural Network(GAT-CNN)to address the aforementioned issues.The main contributions of this paper are as follows:Firstly,we present a three-tiered hierarchical cloud-edge cooperative framework that increases the volume of user feature input by means of two aggregations at the device,edge,and cloud layers.Second,we extend the use of users'own data by introducing self-attention and graph convolution models to track the relative changes of both users and packages simultaneously.Lastly,we build an integrated offline-online system for churn prediction based on the strengths of the two models,and we experimentally validate the efficacy of cloudside collaborative training and inference.In summary,the churn prediction model based on Graph Attention Convolutional Neural Network presented in this paper can effectively address the drawbacks of conventional algorithms and offer telecom operators crucial decision support in developing subscriber retention strategies and cutting operational expenses.

An improved GCN−TCN−AR model for PM_(2.5) predictions in the arid areas of Xinjiang,China

As one of the main characteristics of atmospheric pollutants,PM_(2.5) severely affects human health and has received widespread attention in recent years.How to predict the variations of PM_(2.5) concentrations with high accuracy is an important topic.The PM_(2.5) monitoring stations in Xinjiang Uygur Autonomous Region,China,are unevenly distributed,which makes it challenging to conduct comprehensive analyses and predictions.Therefore,this study primarily addresses the limitations mentioned above and the poor generalization ability of PM_(2.5) concentration prediction models across different monitoring stations.We chose the northern slope of the Tianshan Mountains as the study area and took the January−December in 2019 as the research period.On the basis of data from 21 PM_(2.5) monitoring stations as well as meteorological data(temperature,instantaneous wind speed,and pressure),we developed an improved model,namely GCN−TCN−AR(where GCN is the graph convolution network,TCN is the temporal convolutional network,and AR is the autoregression),for predicting PM_(2.5) concentrations on the northern slope of the Tianshan Mountains.The GCN−TCN−AR model is composed of an improved GCN model,a TCN model,and an AR model.The results revealed that the R2 values predicted by the GCN−TCN−AR model at the four monitoring stations(Urumqi,Wujiaqu,Shihezi,and Changji)were 0.93,0.91,0.93,and 0.92,respectively,and the RMSE(root mean square error)values were 6.85,7.52,7.01,and 7.28μg/m^(3),respectively.The performance of the GCN−TCN−AR model was also compared with the currently neural network models,including the GCN−TCN,GCN,TCN,Support Vector Regression(SVR),and AR.The GCN−TCN−AR outperformed the other current neural network models,with high prediction accuracy and good stability,making it especially suitable for the predictions of PM_(2.5)concentrations.This study revealed the significant spatiotemporal variations of PM_(2.5)concentrations.First,the PM_(2.5) concentrations exhibited clear seasonal fluctuations,with higher levels typically observed in winter and differences presented between months.Second,the spatial distribution analysis revealed that cities such as Urumqi and Wujiaqu have high PM_(2.5) concentrations,with a noticeable geographical clustering of pollutions.Understanding the variations in PM_(2.5) concentrations is highly important for the sustainable development of ecological environment in arid areas.

Adaptively temporal graph convolution model for epidemic prediction of multiple age groups

Introduction:Multivariate time series prediction of infectious diseases is significant to public health,and the deep learning method has attracted increasing attention in this research field.Material and methods:An adaptively temporal graph convolution(ATGCN)model,which leams the contact patterns of multiple age groups in a graph-based approach,was proposed for COVID-19 and influenza prediction.We compared ATGCN with autoregressive models,deep sequence learning models,and experience-based ATGCN models in short-term and long-term prediction tasks.Results:Results showed that the ATGCN model performed better than the autoregressive models and the deep sequence learning models on two datasets in both short-term(12.5%and 10%improvements on RMSE)and longterm(12.4%and 5%improvements on RMSE)prediction tasks.And the RMSE of ATGCN predictions fluctuated least in different age groups of COVID-19(0.029±0.003)and influenza(0.059±0.008).Compared with the Ones-ATGCN model or the Pre-ATGCN model,the ATGCN model was more robust in performance,with RMSE of 0.0293 and 0.06 on two datasets when horizon is one.Discussion:Our research indicates a broad application prospect of deep learning in the field of infectious disease prediction.Transmission characteristics and domain knowledge of infectious diseases should be further applied to the design of deep learning models and feature selection.Conclusion:The ATGCN model addressed the multivariate time series forecasting in a graph-based deep learning approach and achieved robust prediction on the confirmed cases of multiple age groups,indicating its great potentials for exploring the implicit interactions of multivariate variables.

Vehicle Detection Based on Visual Saliency and Deep Sparse Convolution Hierarchical Model

Traditional vehicle detection algorithms use traverse search based vehicle candidate generation and hand crafted based classifier training for vehicle candidate verification.These types of methods generally have high processing times and low vehicle detection performance.To address this issue,a visual saliency and deep sparse convolution hierarchical model based vehicle detection algorithm is proposed.A visual saliency calculation is firstly used to generate a small vehicle candidate area.The vehicle candidate sub images are then loaded into a sparse deep convolution hierarchical model with an SVM-based classifier to perform the final detection.The experimental results demonstrate that the proposed method is with 94.81% correct rate and 0.78% false detection rate on the existing datasets and the real road pictures captured by our group,which outperforms the existing state-of-the-art algorithms.More importantly,high discriminative multi-scale features are generated by deep sparse convolution network which has broad application prospects in target recognition in the field of intelligent vehicle.

Weighted Forwarding in Graph Convolution Networks for Recommendation Information Systems

Recommendation Information Systems(RIS)are pivotal in helping users in swiftly locating desired content from the vast amount of information available on the Internet.Graph Convolution Network(GCN)algorithms have been employed to implement the RIS efficiently.However,the GCN algorithm faces limitations in terms of performance enhancement owing to the due to the embedding value-vanishing problem that occurs during the learning process.To address this issue,we propose a Weighted Forwarding method using the GCN(WF-GCN)algorithm.The proposed method involves multiplying the embedding results with different weights for each hop layer during graph learning.By applying the WF-GCN algorithm,which adjusts weights for each hop layer before forwarding to the next,nodes with many neighbors achieve higher embedding values.This approach facilitates the learning of more hop layers within the GCN framework.The efficacy of the WF-GCN was demonstrated through its application to various datasets.In the MovieLens dataset,the implementation of WF-GCN in LightGCN resulted in significant performance improvements,with recall and NDCG increasing by up to+163.64%and+132.04%,respectively.Similarly,in the Last.FM dataset,LightGCN using WF-GCN enhanced with WF-GCN showed substantial improvements,with the recall and NDCG metrics rising by up to+174.40%and+169.95%,respectively.Furthermore,the application of WF-GCN to Self-supervised Graph Learning(SGL)and Simple Graph Contrastive Learning(SimGCL)also demonstrated notable enhancements in both recall and NDCG across these datasets.

Functionalized Data Operator Model for System Analysis and Forecasting

A deconvolution data processing is developed for obtaining a Functionalized Data Operator (FDO) model that is trained to approximate past and present, input-output data relations. The FDO model is designed to predict future output features for deviated input vectors from any expected, feared of conceivable, future input for optimum control, forecast, or early-warning hazard evaluation. The linearized FDO provides fast analytical, input-output solution in matrix equation form. If the FDO is invertible, the necessary input for a desired output may be explicitly evaluated. A numerical example is presented for FDO model identification and hazard evaluation for methane inflow into the working face in an underground mine: First, a Physics-Based Operator (PBO) model to match monitored data. Second, FDO models are identified for matching the observed, short-term variations with time in the measured data of methane inflow, varying model parameters and simplifications following the parsimony concept of Occam’s Razor. The numerical coefficients of the PBO and FDO models are found to differ by two to three orders of magnitude for methane release as a function of short-time barometric pressure variations. As being data-driven, the significantly different results from an FDO versus PBO model is either an indication of methane release processes poorly understood and modeled in PBO, missing some physics for the pressure spikes;or of problems in the monitored data fluctuations, erroneously sampled with time;or of false correlation. Either way, the FDO model is originated from the functionalized form of the monitored data, and its result is considered experimentally significant within the specified RMS error of model matching.

A Modeling Method Applied to Fault Diagnosis for Constant Linear System

The paper presents a new modeling method applied to fault diagnosis for constant linear closed-loop system by taking the impulse response series as the system model, and provides the calculation process of the method and output of model. The high frequency part of the pulse series, in the method, is reversed so as not to lose the frequency information of the pulse series in its transfer function. On the other hand, the method can also avoid the disadvantage that the learning results of neural network are uncertain every time. In the last part, the application with random disturbance of digital simulation and practical system shows that the modeling method is high accurate and suitable to be applied in fault diagnosis area.

Label-free breast cancer detection and classification by convolutional neural network-based on exosomes surface-enhanced raman scattering

Because the breast cancer is an important factor that threatens women's lives and health,early diagnosis is helpful for disease screening and a good prognosis.Exosomes are nanovesicles,secreted from cells and other body fluids,which can reflect the genetic and phenotypic status of parental cells.Compared with other methods for early diagnosis of cancer(such as circulating tumor cells(CTCs)and circulating tumor DNA),exosomes have a richer number and stronger biological stability,and have great potential in early diagnosis.Thus,it has been proposed as promising biomarkers for diagnosis of early-stage cancer.However,distinguishing different exosomes remain is a major biomedical challenge.In this paper,we used predictive Convolutional Neural model to detect and analyze exosomes of normal and cancer cells with surface-enhanced Raman scattering(SERS).As a result,it can be seen from the SERS spectra that the exosomes of MCF-7,MDA-MB-231 and MCF-10A cells have similar peaks(939,1145 and 1380 cm^(-1)).Based on this dataset,the predictive model can achieve 95%accuracy.Compared with principal component analysis(PCA),the trained CNN can classify exosomes from different breast cancer cells with a superior performance.The results indicate that using the sensitivity of Raman detection and exosomes stable presence in the incubation period of cancer cells,SERS detection combined with CNN screening may be used for the early diagnosis of breast cancer in the future.

Sparse convolutional model with semantic expression for waste electrical appliances recognition

Deep neural networks play an important role in the recognition of waste electrical appliances. However, deep neural network components still lack reliability in decision-making features. To address this problem, a sparse convolutional model with semantic expression(SCMSE) is proposed. First, a low-rank sparse semantic expression component, combining the benefits of residual networks and sparse representation, is adapted to enhance sparse feature extraction and semantic expression. Second, a reliable network architecture is obtained by iterating the optimal sparse solution, enhancing semantic expression. Finally, the results of visualization experiments on the waste electrical appliances dataset demonstrate that the proposed SCMSE can obtain excellent semantic performance.

Multi-Layer Deep Sparse Representation for Biological Slice Image Inpainting

Biological slices are an effective tool for studying the physiological structure and evolutionmechanism of biological systems.However,due to the complexity of preparation technology and the presence of many uncontrollable factors during the preparation processing,leads to problems such as difficulty in preparing slice images and breakage of slice images.Therefore,we proposed a biological slice image small-scale corruption inpainting algorithm with interpretability based on multi-layer deep sparse representation,achieving the high-fidelity reconstruction of slice images.We further discussed the relationship between deep convolutional neural networks and sparse representation,ensuring the high-fidelity characteristic of the algorithm first.A novel deep wavelet dictionary is proposed that can better obtain image prior and possess learnable feature.And multi-layer deep sparse representation is used to implement dictionary learning,acquiring better signal expression.Compared with methods such as NLABH,Shearlet,Partial Differential Equation(PDE),K-Singular Value Decomposition(K-SVD),Convolutional Sparse Coding,and Deep Image Prior,the proposed algorithm has better subjective reconstruction and objective evaluation with small-scale image data,which realized high-fidelity inpainting,under the condition of small-scale image data.And theOn2-level time complexitymakes the proposed algorithm practical.The proposed algorithm can be effectively extended to other cross-sectional image inpainting problems,such as magnetic resonance images,and computed tomography images.

Three-dimensional temperature reconstruction of diffusion flame from the light-field convolution imaging by the focused plenoptic camera

The plenoptic imaging technique provides a promising approach to the non-invasive three-dimensional measurement, especially for the high-temperature combustion diagnosis. We establish a light-field convolution imaging model for diffusion flame in this work, considering the radiation transfer process inside the diffusion flame and the light transfer process inside the focused plenoptic camera together. The radiation transfer process is described by the radiation transfer equation and solved by the generalized source multi-flux method. Wave optics theory is adopted to describe the light transfer process, combining Fresnel diffraction and the phase conversion of the lens. The flame light-field image is obtained by the light-field convolution imaging model and adopted as the measurement signal to reconstruct three-dimensional temperature field. The inverse problem of temperature reconstruction is solved by the least square QR decomposition method. The simulative temperature reconstruction work is conducted, including the inverse analysis, the uncertainty analysis, and the measurement noise influence. All the results show that the proposed measurement method is available to reconstruct three-dimensional temperature with satisfactory accuracy and acceptable uncertainty. Both symmetric and asymmetric distributed temperature fields are investigated, and the reconstructed results prove the validity and universality of the measurement method.

Automatic segmentation algorithm for high-spatial-resolution remote sensing images based on self-learning super-pixel convolutional network

Super-pixel algorithms based on convolutional neural networks with fuzzy C-means clustering are widely used for high-spatial-resolution remote sensing images segmentation.However,this model requires the number of clusters to be set manually,resulting in a low automation degree due to the complexity of the iterative clustering process.To address this problem,a segmentation method based on a self-learning super-pixel network(SLSP-Net)and modified automatic fuzzy clustering(MAFC)is proposed.SLSP-Net performs feature extraction,non-iterative clustering,and gradient reconstruction.A lightweight feature embedder is adopted for feature extraction,thus expanding the receiving range and generating multi-scale features.Automatic matching is used for non-iterative clustering,and the overfitting of the network model is overcome by adaptively adjusting the gradient weight parameters,providing a better irregular super-pixel neighborhood structure.An optimized density peak algorithm is adopted for MAFC.Based on the obtained super-pixel image,this maximizes the robust decision-making interval,which enhances the automation of regional clustering.Finally,prior entropy fuzzy C-means clustering is applied to optimize the robust decision-making and obtain the final segmentation result.Experimental results show that the proposed model offers reduced experimental complexity and achieves good performance,realizing not only automatic image segmentation,but also good segmentation results.

A Software Defect Prediction Method Using a Multivariate Heterogeneous Hybrid Deep Learning Algorithm

Software defect prediction plays a critical role in software development and quality assurance processes. Effective defect prediction enables testers to accurately prioritize testing efforts and enhance defect detection efficiency. Additionally, this technology provides developers with a means to quickly identify errors, thereby improving software robustness and overall quality. However, current research in software defect prediction often faces challenges, such as relying on a single data source or failing to adequately account for the characteristics of multiple coexisting data sources. This approach may overlook the differences and potential value of various data sources, affecting the accuracy and generalization performance of prediction results. To address this issue, this study proposes a multivariate heterogeneous hybrid deep learning algorithm for defect prediction (DP-MHHDL). Initially, Abstract Syntax Tree (AST), Code Dependency Network (CDN), and code static quality metrics are extracted from source code files and used as inputs to ensure data diversity. Subsequently, for the three types of heterogeneous data, the study employs a graph convolutional network optimization model based on adjacency and spatial topologies, a Convolutional Neural Network-Bidirectional Long Short-Term Memory (CNN-BiLSTM) hybrid neural network model, and a TabNet model to extract data features. These features are then concatenated and processed through a fully connected neural network for defect prediction. Finally, the proposed framework is evaluated using ten promise defect repository projects, and performance is assessed with three metrics: F1, Area under the curve (AUC), and Matthews correlation coefficient (MCC). The experimental results demonstrate that the proposed algorithm outperforms existing methods, offering a novel solution for software defect prediction.

基于双支路特征融合的MRI颅脑肿瘤图像分割研究

针对磁共振成像(magnetic resonance imaging, MRI)颅脑肿瘤区域误识别与分割网络空间信息丢失问题,提出一种基于双支路特征融合的MRI脑肿瘤图像分割方法。首先通过主支路的重构VGG与注意力模型(re-parameterization visual geometry group and attention model, RVAM)提取网络的上下文信息,然后使用可变形卷积与金字塔池化模型(deformable convolution and pyramid pooling model, DCPM)在副支路获取丰富的空间信息,之后使用特征融合模块对两支路的特征信息进行融合。最后引入注意力模型,在上采样过程中加强分割目标在解码时的权重。提出的方法在Kaggle_3m数据集和BraTS2019数据集上进行了实验验证,实验结果表明该方法具有良好的脑肿瘤分割性能,其中在Kaggle_3m上,Dice相似系数、杰卡德系数分别达到了91.45%和85.19%。

Maximizing the mechanical performance of Ti_(3)AlC_(2)-based MAX phases with aid of machine learning

Mechanical properties consisting of the bulk modulus,shear modulus,Young’s modulus,Poisson’s ratio,etc.,are key factors in determining the practical applications of MAX phases.These mechanical properties are mainly dependent on the strength of M–X and M–A bonds.In this study,a novel strategy based on the crystal graph convolution neural network(CGCNN)model has been successfully employed to tune these mechanical properties of Ti_(3)AlC_(2)-based MAX phases via the A-site substitution(Ti_(3)(Al1-xAx)C_(2)).The structure–property correlation between the A-site substitution and mechanical properties of Ti_(3)(Al1-xAx)C_(2)is established.The results show that the thermodynamic stability of Ti_(3)(Al1-xAx)C_(2)is enhanced with substitutions A=Ga,Si,Sn,Ge,Te,As,or Sb.The stiffness of Ti_(3)AlC_(2)increases with the substitution concentration of Si or As increasing,and the higher thermal shock resistance is closely associated with the substitution of Sn or Te.In addition,the plasticity of Ti_(3)AlC_(2)can be greatly improved when As,Sn,or Ge is used as a substitution.The findings and understandings demonstrated herein can provide universal guidance for the individual synthesis of high-performance MAX phases for various applications.