Feature Matrix Research Articles

Advanced method for recognizing and measuring key information in non-stationary signals using variational mode decomposition and CNN-LSTM-SVM algorithms.

In order to address the issue of insufficient ability to identify and measure the key information of non-stationary signals collected in practical industrial fields such as logistics transportation, state detection, and fault diagnosis, this paper proposes a method to identify and measure the key information based on variational mode decomposition (VMD), convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and support vector machine (SVM). First, the non-stationary signal is reconstructed by using VMD and linear correlation decomposition. Second, the feature matrix is constructed according to the upper envelope feature, moving kurtosis, and moving root mean square. Finally, CNN-LSTM-SVM is input to identify and measure the key features. The experimental results demonstrate that the proposed method exhibits an outstanding performance on both synthetic and actual collected signals, with recognition accuracies of 99.17% and 99.02%, respectively.

Enhancing deep stochastic configuration networks: Efficient training via low-rank matrix approximation

Deep Stochastic Configuration Networks (DeepSCN) represent a class of randomized learner models distinguished by their universal approximation capabilities, rapid learning processes, and ease of implementation. These models surpass other randomized learners due to an embedded supervisory mechanism. Although the feature matrix is meticulously constructed, the predictive accuracy of DeepSCN also critically depends on the quality of the model parameters obtained from solving the corresponding system of linear equations. As the number of hidden neurons increases, the resulting expansion of the feature matrix renders matrix inversion computationally expensive, potentially unstable, and memory-intensive. This study addresses these challenges by integrating randomized algorithm-based low-rank matrix approximation to compute model parameters efficiently. A theoretical error bound for the obtained model parameters is derived, ensuring precision and accuracy. The effectiveness of the algorithm is validated on benchmark datasets for both classification and regression tasks. The results demonstrate that this integration significantly enhances the reliability and stability of DeepSCN, improving performance and scalability while greatly reducing computational demands. The method maintains stability despite substantial variations in the number of hidden neurons and mitigates overfitting caused by matrix inversion, ensuring stable and reliable training outcomes. This study advances DeepSCN training methodologies and large-scale data processing.

LncRNA-miRNA interactions prediction based on meta-path similarity and Gaussian kernel similarity.

Long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) are two typical types of non-coding RNAs that interact and play important regulatory roles in many animal organisms. Exploring the unknown interactions between lncRNAs and miRNAs contributes to a better understanding of their functional involvement. Currently, studying the interactions between lncRNAs and miRNAs heavily relies on laborious biological experiments. Therefore, it is necessary to design a computational method for predicting lncRNA-miRNA interactions. In this work, we propose a method called MPGK-LMI, which utilizes a graph attention network (GAT) to predict lncRNA-miRNA interactions in animals. First, we construct a meta-path similarity matrix based on known lncRNA-miRNA interaction information. Then, we use GAT to aggregate the constructed meta-path similarity matrix and the computed Gaussian kernel similarity matrix to update the feature matrix with neighbourhood information. Finally, a scoring module is used for prediction. By comparing with three state-of-the-art algorithms, MPGK-LMI achieves the best results in terms of performance, with AUC value of 0.9077, AUPR of 0.9327, ACC of 0.9080, F1-score of 0.9143 and precision of 0.8739. These results validate the effectiveness and reliability of MPGK-LMI. Additionally, we conduct detailed case studies to demonstrate the effectiveness and feasibility of our approach in practical applications. Through these empirical results, we gain deeper insights into the functional roles and mechanisms of lncRNA-miRNA interactions, providing significant breakthroughs and advancements in this field of research. In summary, our method not only outperforms others in terms of performance but also establishes its practicality and reliability in biological research through real-case analysis, offering strong support and guidance for future studies and applications.

EEG FEATURE EXTRACTION AND RECOGNITION BASED ON MULTI-FEATURE FUSION

In response to the issue of insufficient information and low classification accuracy in single-channel feature extraction algorithms for EEG signals, the paper proposed a method based on Adaptive Noise-assisted Complete Ensemble Empirical Mode Decomposition (CEEMDAN) and multi-feature fusion. Firstly, we acquired multi-channel motor imagery EEG data and preprocessed it. Next, utilized CEEMDAN with adaptive noise to adaptively decompose the channel data. Secondly, extracted features from the Intrinsic Mode Functions (IMFs) which we obtained after CEEMDAN decomposition for different feature layers and fused them, obtaining the original feature matrices for each feature layer. Finally, we fused and reduced the dimensionality of the original feature matrices from each layer, input the resulting matrix into a classification model for classification, and output the classification results. We achieve an average classification accuracy of 88.59% on the BCI Competition II Data Set III, and validate the results on self-collected single-channel motor imagery EEG data, achieving a classification accuracy of 89.72%. Experimental results indicate that multi-feature fusion outperforms single features, and the combination of CEEMDAN decomposition with multi-feature fusion achieves higher classification accuracy.

Development and optimization of an Artificial Neural Network (ANN) model for predicting the cadmium fixation efficiency of biochar in soil

This study addresses the issue of soil cadmium pollution and the common problem of data missing during the training of artificial neural network models. A substantial amount of characteristic data regarding the immobilization of cadmium in soil through biochar was collected from the literature. Subsequently, data was organized and analyzed using correlation matrix analysis and feature importance analysis. The results revealed that the feature importance of biochar (52%) was relatively significant. A novel approach is introduced, involving a non-deterministic optimization technique based on linear interpolation to supplement the missing data within the dataset. The completed dataset effectively preserves the distributional characteristics of the original data, thereby enhancing the training performance of the neural network model. Through three phases, namely data incompleteness, data supplementation, and genetic algorithm optimization, the neural network model was progressively refined. The outcome of this iterative process is a high-performance artificial neural network model capable of predicting the efficiency of biochar in immobilizing cadmium in soil. The significance of these findings lies in their potential to advance the application of machine learning in environmental science, particularly in addressing complex pollution scenarios.

Chemical and mineralogical constitution of redoximorphic features and mechanism of formation of Plinthosols from the Araguaia River plain, Brazil

ABSTRACT Currently in Brazil, large grain cultivation projects on Plinthosols are a reality, however, there is little or no knowledge of the real mechanism of formation of the plinthite feature, in addition to what is reported in the literature as being a product of oxidation-reduction processes of iron element. This study evaluates iron redoximorphic features and investigates their chemical and mineralogical composition in two profiles of Plinthosols from the Araguaia River plain (P1 and P2). The study strengthens the understanding of the pedogenetic processes involved in the formation of mottles and plinthite. In this sense, it assesses whether the formation mechanisms corroborate the literature. Soil features were sampled in the upper right and left position at the initial plinthic horizon, upper right and left position at the main plinthic horizon, and lower right position at the base horizon of the plinthite zone in the profile. Separated samples comprising the soil matrix, mottles, and plinthite under natural moisture conditions were ground into powder form for chemical determinations by X-ray fluorescence (XRF), sulfuric acid attack (H 2 SO 4 ), sodium dithionite-citrate-bicarbonate (DCB), and ammonium acid oxalate; and mineralogical determinations by X-ray diffraction. Iron contents in all determined forms were always higher in the plinthite feature, intermediate in the mottle feature, and lower in the soil matrix feature. Most of the Fe in all redoximorphic features is included in the structure of primary minerals and their derivatives (vermiculite, illite, and VHEs). Only part of the iron present (about 35.40 % in P1 and 41.98 % in P2) is detected in the form of oxides such as goethite and hematite, which could be formed in redox processes. The mottle and plinthite features under study are not the product of the classic process of segregation, mobilization, and accumulation of iron as a consequence of redox processes. These features were formed or emerged as a result of a relatively slow and constant weathering process of their source material, which is gradually decomposed in an aqueous medium, releasing most of its components. These components include iron and more mobile elements such as bases and silicon, which leave the system through drainage water, and of which a small part may eventually recombine to form new less complex minerals such as kaolinite and oxides.

Fault Diagnosis for Motor Bearings via an Intelligent Strategy Combined with Signal Reconstruction and Deep Learning

To overcome the incomplete decomposition of vibration signals in traditional motor-bearing fault diagnosis algorithms and improve the ability to characterize fault characteristics and anti-interference, a diagnostic strategy combining dual signal reconstruction and deep learning architecture is proposed. In this study, an improved complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and variational mode decomposition (VMD)-based signal reconstruction method is first introduced to extract features representing motor bearing faults. A feature matrix construction method based on improved information entropy is then proposed to quantify these fault features. Finally, a fault diagnosis algorithm architecture integrating a multi-scale convolutional neural network (MSCNN) with attention mechanisms and a bidirectional long short-term memory network (BiLSTM) is developed. The experimental results for four fault states show that this model can effectively extract fault features from original vibration signals and, compared to other fault diagnosis models, offer high diagnostic accuracy and strong generalization, maintaining high accuracy even under varying speeds and noise interference.

Multi-dimensional weighted deep subspace clustering with feature classification

Deep subspace clustering (DSC) methods based on deep autoencoder (DAE) and self-expression layers have achieved impressive performance. However, traditional DSC method often loses useful information in the feature extraction of DAE, leading to incomplete learning of the self-expression coefficient matrix. Besides, existing deep subspace clustering methods always extract feature information from a single dimension, ignoring the potential information that may exist in the self-expression coefficient matrix across multiple dimensions. In this paper, we propose a novel method called multi-dimensional weighted deep subspace clustering based on feature classification (MWDSC). Specifically, we use the multivariate feature classification module (MFC) to learn the potential information of multiple local feature matrix for local pre-clustering. To further improve the clustering performance, we propose a self-flip weighted fusion strategy (SWF). The SWF strategy obtains a self-expression coefficient matrix with rich features through self-flipping weighting, thereby aiding in achieving global optimal solutions. Extensive experiments conducted on five benchmark datasets validate the superiority and effectiveness of our proposed method over other state-of-the-art approaches.

SCAE: Structural Contrastive Auto-Encoder for Incomplete Multi-View Representation Learning

Describing an object from multiple perspectives often leads to incomplete data representation. Consequently, learning consistent representations for missing data from multiple views has emerged as a key focus in the realm of Incomplete Multi-view Representation Learning (IMRL). In recent years, various strategies, such as subspace learning, matrix decomposition, and deep learning, have been harnessed to develop numerous IMRL methods. In this article, our primary research revolves around IMRL, with a particular emphasis on addressing two main challenges. Firstly, we delve into the effective integration of intra-view similarity and contextual structure into a unified framework. Secondly, we explore the effective facilitation of information exchange and fusion across multiple views. To tackle these issues, we propose a deep learning approach known as Structural Contrastive Auto-Encoder (SCAE) to solve the challenges of IMRL. SCAE comprises two major components: intra-view structural representation learning and inter-view contrastive representation learning. The former involves capturing intra-view similarity by minimizing the Dirichlet energy of the feature matrix, while also applying spatial dispersion regularization to capture intra-view contextual structure. The latter encourages maximizing the mutual information of inter-view representations, facilitating information exchange and fusion across views. Experimental results demonstrate the efficacy of our approach in significantly enhancing model accuracy and robustly addressing IMRL problems. The code is available at https://github.com/limengran98/SCAE .

Unsupervised clustering of mining-induced microseismicity provides insights into source mechanisms

Microseismic source mechanisms in underground mines can provide information about the rock mass response to mining. Conventional approaches to such studies rely upon moment tensor solutions that are susceptible to modeling assumptions and need reliable information about source locations and high-resolution velocity models. We propose the application of unsupervised clustering to group microseismic events into different classes directly from the waveform data such that the events in a specific class have similar source mechanisms. Our method has three main steps, first using spectral decomposition to separate the source terms from the path-receiver contributions in the observed amplitude spectra of events occurring in spatially dense clusters. Second, reducing the number of features from the source spectra using independent component analysis (ICA). Third, applying a Gaussian mixture model (GMM) to the reduced feature matrix to obtain event clusters. To test our method, we generate synthetic waveform data using the receiver network and the recorded microseismic event locations in an underground potash mine in Saskatchewan. Results show the ability of our method to separate events into different classes corresponding to differences in source mechanisms. Application to field data recorded in the mine during February 2021 successfully discriminates between blasts and microseismic events. The data recorded between 1 March and 30 June 2021 that contain microseismic events only are divided into two dominant classes. Using known moment tensors (MT) of some of these events for labeling, we interpret one of the two classes as having dominant double-couple mechanisms as compared to the other which most likely corresponds to the linear dipole-tensile mechanisms. Our method, combined with some expert knowledge such as MT of some larger magnitude events, can offer an assessment of source types of large microseismic populations as often encountered in induced seismicity.

A Method for Estimating the SOH of Lithium-Ion Batteries Based on Graph Perceptual Neural Network

The accurate estimation of battery state of health (SOH) is critical for ensuring the safety and reliability of devices. Considering the variation in health degradation across different types of lithium-ion battery materials, this paper proposes an SOH estimation method based on a graph perceptual neural network, designed to adapt to multiple battery materials. This method adapts to various battery materials by extracting crucial features from current, voltage, voltage–capacity, and temperature data, and it constructs a graph structure to encapsulate these features. This approach effectively captures the complex interactions and dependencies among different battery types. The novel technique of randomly removing features addresses feature redundancy. Initially, a mutual information graph structure is defined to illustrate the interdependencies among battery features. Moreover, a graph perceptual self-attention mechanism is implemented, integrating the adjacency matrix and edge features into the self-attention calculations. This enhancement aids the model’s understanding of battery behaviors, thereby improving the transparency and interpretability of predictions. The experimental results demonstrate that this method outperforms traditional models in both accuracy and generalizability across various battery types, particularly those with significant chemical and degradation discrepancies. The model achieves a minimum mean absolute error of 0.357, a root mean square error of 0.560, and a maximum error of 0.941.

Peritumoral Adipose Tissue Features Derived from [18F]fluoro-2-deoxy-2-d-glucose Positron Emission Tomography/Computed Tomography as Predictors for Response to Neoadjuvant Chemotherapy in Breast Cancer Patients.

This study investigated whether the textural features of peritumoral adipose tissue (AT) on [18F]fluoro-2-deoxy-2-d-glucose (FDG) positron emission tomography/computed tomography (PET/CT) can predict the pathological response to neoadjuvant chemotherapy (NAC) and progression-free survival (PFS) in breast cancer patients. We retrospectively enrolled 147 female breast cancer patients who underwent staging FDG PET/CT and completed NAC and underwent curative surgery. We extracted 10 first-order features, 6 gray-level co-occurrence matrix (GLCM) features, and 3 neighborhood gray-level difference matrix (NGLDM) features of peritumoral AT and evaluated the predictive value of those imaging features for pathological complete response (pCR) and PFS. The results of our study demonstrated that GLCM homogeneity showed the highest predictability for pCR among the peritumoral AT imaging features in the receiver operating characteristic curve analysis. In multivariate logistic regression analysis, the mean standardized uptake value (SUV), 50th percentile SUV, 75th percentile SUV, SUV histogram entropy, GLCM entropy, and GLCM homogeneity of the peritumoral AT were independent predictors for pCR. In multivariate survival analysis, SUV histogram entropy and GLCM correlation of peritumoral AT were independent predictors of PFS. Textural features of peritumoral AT on FDG PET/CT could be potential imaging biomarkers for predicting the response to NAC and disease progression in breast cancer patients.

GLNNMDA: a multimodal prediction model for microbe-drug associations based on global and local features

Microbes have been demonstrated to be closely linked to diseases that pose a major threat to human health. Computing technologies can help researchers find potential microbe-drug associations more quickly and precisely. In this study, we introduced a novel computational prediction model called GLNNMDA based on global and local features of microbes and drugs to infer possible microbe-drug correlations. In GLNNMDA, we first constructed a heterogeneous network based on known microbe-drug relationships by integrating multiple similarity metrics of drugs and microbes. Subsequently, low-dimensional features will be extracted for nodes in the heterogeneous network by adopting the graph attention encoder. Next, based on combining these low-dimensional features with multiple properties of microbes and drugs to form a new comprehensive feature matrix, we would utilize the GLF module to extract the global and local features for microbes and drugs respectively, and then, we would further fuse these global and local features to come up with predictions of possible microbe-drug associations. Moreover, in order to evaluate the prediction performance of GLNNMDA, under the framework of fivefold cross-validation, intensive comparative experiments and case studies were done on different well-known public databases. The results showed that GLNNMDA obtained the highest AUC values as well as AUPR values of 0.9802 ± 0.0011, 0.9773 ± 0.0021 and 0.8586 ± 0.0004, 0.8008 ± 0.0031 in the two databases, MDAD and aBiofilm, respectively, compared to the state-of-the-art competing prediction methods. In addition, case studies of well-known microorganisms and drugs have demonstrated the effectiveness of GLNNMDA in inferring potential microbial drug associations, which implies that GLNNMDA may be a useful tool for microbe-drug association prediction in the future. The source code is available at: “https://github.com/KuangHaiYue/GLNNMDA.git”.

Frequency distribution-aware network based on discrete cosine transformation (DCT) for remote sensing image super resolution.

Single-image super-resolution technology based on deep learning is widely used in remote sensing. The non-local feature reflects the correlation information between different regions. Most neural networks extract various non-local information of images in the spatial domain but ignore the similarity characteristics of frequency distribution, which limits the performance of the algorithm. To solve this problem, we propose a frequency distribution aware network based on discrete cosine transformation for remote sensing image super-resolution. This network first proposes a frequency-aware module. This module can effectively extract the similarity characteristics of the frequency distribution between different regions by rearranging the frequency feature matrix of the image. A global frequency feature fusion module is also proposed. It can extract the non-local information of feature maps at different scales in the frequency domain with little computational cost. The experiments were on two commonly-used remote sensing datasets. The experimental results show that the proposed algorithm can effectively complete image reconstruction and performs better than some advanced super-resolution algorithms. The code is available at https://github.com/Liyszepc/FDANet.

A Hybrid Segmentation Algorithm for Rheumatoid Arthritis Diagnosis Using X-ray Images

Rheumatoid Arthritis (RA) is a chronic autoimmune illness that occurs in the joints, resulting in inflammation, pain, and stiffness. X-ray examination is one of the most common diagnostic procedures for RA, but manual X-ray image analysis has limitations because it is a time-consuming procedure and is prone to errors. A specific algorithm aims to a lay stable and accurate segmenting of carpal bones from hand bone images, which is vitally important for identifying rheumatoid arthritis. The algorithm demonstrates several stages, starting with Carpal bone Region of Interest (CROI) specification, dynamic thresholding, and Gray Level Co-occurrence Matrix (GLCM) application for texture analysis. To get the clear edges of the image, the component is first converted to the greyscale function and thresholding is carried out to separate the hand from the background. The pad region is identified to obtain the contours of it, and the CROI is defined by the bounding box of the largest contour. The threshold value used in the CROI method is given a dynamic feature that can separate the carpal bones from the surrounding tissue. Then the GLCM texture analysis is carried out, calculating the number of pixel neighbors, with the specific intensity and neighbor relations of the pixels. The resulting feature matrix is then employed to extract features such as contrast and energy, which are later used to categorize the images of the affected carpal bone into inflamed and normal. The proposed technique is tested on a rheumatoid arthritis image dataset, and the results show its contribution to diagnosis of the disease. The algorithm efficiently divides carpal bones and extracts the signature parameters that are critical for correct classification of the inflammation in the cartilage images.

Reimagining otitis media diagnosis: A fusion of nested U-Net segmentation with graph theory-inspired feature set

Otitis media (OM) is a common infection or inflammation of the middle ear causing conductive hearing loss that primarily affects children and may delay speech, language, and cognitive development. OM can manifest itself in different forms, and can be diagnosed using (video) otoscopy (visualizing the tympanic membrane) or (video) pneumatic otoscopy and tympanometry. Accurate diagnosis of OM is challenging due to subtle differences in otoscopic features. This research aims to develop an automated computer-aided design (CAD) system to assist clinicians in diagnosing OM using otoscopy images. The ground truths, generated manually and validated by otolaryngologists, are utilized to train the proposed nested U-Net++ model. Ten clinically relevant gray level co-occurrence matrix (GLCM) and morphological features were extracted from the segmented Region of Interest (ROI) and validated for OM classification based on a statistical significance test. These features serve as input for a Graph Neural Network (GNN) model, the base model in our research. An optimized GNN model is proposed after ablation study of the base model. Three datasets, one private dataset, and two public ones have been used, where the private dataset is utilized for both training and testing, and the public datasets are used to test the robustness of the proposed GNN model only. The proposed GNN model obtained the highest accuracy in diagnosing OM: 99.38 %, 93.51 %, and 91.38 % for the private dataset, public dataset1, and public dataset2, respectively. The proposed methodology and results of this research might enhance clinicians' effectiveness in diagnosing OM.

Fibrotic liver extracellular matrix induces cancerous phenotype in biomimetic micro-tissues of hepatocellular carcinoma model

BackgroundDespite considerable advancements in identifying factors contributing to the development of hepatocellular carcinoma (HCC), the pathogenesis of HCC remains unclear. In many cases, HCC is a consequence of prolonged liver fibrosis, resulting in the formation of an intricate premalignant microenvironment. The accumulation of extracellular matrix (ECM) is a hallmark of premalignant microenvironment. Given the critical role of different matrix components in regulating cell phenotype and function, this study aimed to elucidate the interplay between the fibrotic matrix and malignant features in HCC. MethodsLiver tissues from both control (normal) and carbon tetrachloride (CCl4)-induced fibrotic rats were decellularized using sodium dodecyl sulfate (SDS) and Triton X-100. The resulting hydrogel from decellularized ECM was processed into micro-particles via the water-in-oil emulsion method. Micro-particles were subsequently incorporated into three-dimensional liver biomimetic micro-tissues (MTs) comprising Huh-7 cells, human umbilical vein endothelial cells (HUVECs), and LX-2 cells. The MTs were evaluated using the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay at day 11, immunofluorescence staining, immunoblotting, and spheroid migration assay at day 14 after co-culture. ResultsFibrotic matrix from CCl4-treated rat livers significantly enhanced the growth rate of the MTs and their expression of CCND1 as compared to the normal one. Fibrotic matrix, also induced the expression of epithelial-to-mesenchymal transition (EMT)-associated genes such as TWIST1, ACTA2, MMP9, CDH2, and VIMENTIN in the MTs as compared to the normal matrix. Conversely, the expression of CDH1 and hepatic maturation genes HNF4A, ALB, CYP3A4 was decreased in the MTs when the fibrotic matrix was used. Furthermore, the fibrotic matrix increased the migration of the MTs and their secretion of alpha-fetoprotein. ConclusionsOur findings suggest a regulatory role for the fibrotic matrix in promoting cancerous phenotype, which could potentially accelerate the progression of malignancy in the liver.

Relative position matrix and multi-scale feature fusion for writer-independent online signature verification

Online signature verification (OSV) is widely used in finance, law and other fields, and is one of the important research projects on biological characteristics. However, its data set has a small scale and has high requirements for generalization of certification models. Therefore, how to overcome these problems is of great value to improve the practicality and security of online handwriting signature technology. We propose a writer-independent online handwritten signature verification method, which adopts the relative position matrix method to convert the traditional temporal features into images for processing. This method enriched the features of the signatures, serving the purpose of data augmentation. Then two-dimensional multi-scale feature fusion based Siamese neural network (2D-MFFnet) is built for representing and learning the importance of each channel adaptively combined with the attention mechanism. Finally, a temporal convolutional network is designed to construct the classifier. The results illustrate that compared with traditional time series models, the algorithm has reduced the equal error rate by at least 2.52 % on the open datasets MCYT-100 and SVC2004 task2.

Analysis of Cancer-Associated Mutations of POLB Using Machine Learning and Bioinformatics.

DNA damage is a critical factor in the onset and progression of cancer. When DNA is damaged, the number of genetic mutations increases, making it necessary to activate DNA repair mechanisms. A crucial factor in the base excision repair process, which helps maintain the stability of the genome, is an enzyme called DNA polymerase β (Pol β) encoded by the POLB gene. It plays a vital role in the repair of damaged DNA. Additionally, variations known as Single Nucleotide Polymorphisms (SNPs) in the POLB gene can potentially affect the ability to repair DNA. This study uses bioinformatics tools that extract important features from SNPs to construct a feature matrix, which is then used in combination with machine learning algorithms to predict the likelihood of developing cancer associated with a specific mutation. Eight different machine learning algorithms were used to investigate the relationship between POLB gene variations and their potential role in cancer onset. This study not only highlights the complex link between POLB gene SNPs and cancer, but also underscores the effectiveness of machine learning approaches in genomic studies, paving the way for advanced predictive models in genetic and cancer research.

A novel structural deformation prediction method based on graph convolutional network during shield tunnel construction

During shield tunneling through existing steel reinforced concrete structures, superstructure deformation is an important parameter that reflects the disturbance degree of engineering construction to existing structure. Precisely predicting structural deformation can help engineers adjust shield machine operational parameters and ensure the success of the project. There has been no attempt to study the feasibility and applicability of machine learning for predicting structural deformation when shield machine cut through existing structure. To address this problem, this paper proposes a novel hybrid model (DSGCN-TCN), combining dynamic spatial graph convolutional network (DSGCN) and temporal convolutional network (TCN), to predict structural deformation. First, dynamic adjacency matrix is constructed based on correlation coefficient and attention mechanism to describe the dynamic change of irregular graph structure. Then dynamic adjacency matrices and feature matrices as the input of the GCN model to extract the dynamic spatial feature of structural deformation data. Followed by TCN and attention layer to capture the temporal correlation of structural deformation data. Finally, the prediction performance of the proposed method is verified using measured data from practical engineering. The experiment results show that compared with the selected baseline models and sub-models, the proposed model can predict the structural deformation more accurately.