Graph Convolutional Network Research Articles

Data augmentation framework with enhanced graph convolutional network for cross-domain aspect-based sentiment analysis

Data augmentation framework with enhanced graph convolutional network for cross-domain aspect-based sentiment analysis

TEA-GCN: Transformer-Enhanced Adaptive Graph Convolutional Network for Traffic Flow Forecasting.

Traffic flow forecasting is crucial for improving urban traffic management and reducing resource consumption. Accurate traffic conditions prediction requires capturing the complex spatial-temporal dependencies inherent in traffic data. Traditional spatial-temporal graph modeling methods often rely on fixed road network structures, failing to account for the dynamic spatial correlations that vary over time. To address this, we propose a Transformer-Enhanced Adaptive Graph Convolutional Network (TEA-GCN) that alternately learns temporal and spatial correlations in traffic data layer-by-layer. Specifically, we design an adaptive graph convolutional module to dynamically capture implicit road dependencies at different time levels and a local-global temporal attention module to simultaneously capture long-term and short-term temporal dependencies. Experimental results on two public traffic datasets demonstrate the effectiveness of the proposed model compared to other state-of-the-art traffic flow prediction methods.

The study of Alzheimer's disease risk diagnosis based on natural killer cell marker genes in the multi-omics data.

Alzheimer's disease (AD) is a common neurodegenerative disorder, currently lacking effective early diagnostic methods. However, natural killer (NK) cells may play a potential role in AD pathogenesis. This study aims to identify AD-related feature genes from NK cell markers to construct a diagnostic model and explore their potential biological mechanisms in AD. Single-cell RNA sequencing data was used to identify NK cell markers. A novel feature selection algorithm, adaptive dynamic graph convolutional network (ADGCN), was proposed to extract AD-related feature genes and construct a diagnostic model. Differential, correlation and enrichment analyses were performed to understand the biological mechanisms of these genes. Immune infiltration analysis compared the immune microenvironment between AD and controls. Two regulatory networks explored interactions between feature genes, transcription factors and microRNAs. The association between SNPs and feature genes' expression was examined through expression quantitative trait loci analysis. Differential CpG sites were identified to analyze their association with the NK cell markers' expression. We developed an optimal diagnostic model (ADGCN-XGBoost) with 17 feature genes, demonstrating high diagnostic effectiveness across datasets. These genes were primarily related to macromolecule biosynthesis, cytoplasmic translation biological processes and ribosome pathway, and potentially modulated immune infiltration of AD patients. We predicted 27 target miRNAs and 21 transcription factors influencing these genes. Multimodal analysis identified 57 significant SNP-gene associations and seven CpG-gene pairs. This study proposed a novel feature selection algorithm and developed a diagnostic model based on 17 feature genes, providing new potential biomarkers for AD diagnosis.

Tfgcn: a time-varying fuzzy graph convolutional network for multi-sensor traffic flow forecasting

Tfgcn: a time-varying fuzzy graph convolutional network for multi-sensor traffic flow forecasting

A coarse aggregate particle size classification method by fusing 3D multi‐view and graph convolutional networks

A coarse aggregate particle size classification method by fusing 3D multi‐view and graph convolutional networks

Comparative analysis of matrix factorization and graph convolutional networks in student

Abstract. For the educational worker, predicting students grades and having a deep understanding of students levels is quite important to improve their teaching methods. Fortunately, there have been several research to predict students grades in applying different models, such as Matrix Factorization (MF) and Graph Convolutional Networks (GCNs). This essay is talking about comparing two different models, MF and GCNs, which are going to exhibit the difference between them. By comparing their performance in predictive accuracy, interpretability, and computational efficiency, people can identify their strengths and areas for improvement. In this essay, the advantages and disadvantages of the two models will be listed and their performance will be compared. Therefore, in this context, this essay will introduce two models first, then show their performance in different experiences from past research and compare their results. As a result, MF shows a better performance in handling large-scale sparse datasets and providing meaningful interpretations, whereas GCNs are good at capturing complex dependencies and integrating multiple data sources.

Boosting point cloud understanding through graph convolutional network with scale measurement and high-frequency enhancement

Graph-based methods have exhibited exceptional performance in point cloud understanding by capturing local geometric relationships. However, existing approaches often struggle to characterize the overall spatial scale of local graphs. In addition, they fail to capture the differences between nodes effectively, which is crucial for distinguishing different classes. This study introduces SM-HFEGCN, a novel graph convolutional network that addresses these limitations through two key innovations: scale measurement and high-frequency enhancement. First, we introduce a spatial scale feature derived from the diagonal vectors of the neighborhood, which serves as a unique graph-specific property related to the geometry and density of the local point cloud. This feature can characterize the overall spatial scale of the local point cloud. Second, we enhance the high-frequency information to capture node variations and integrate it with smoothed information to represent the differences and similarities between nodes simultaneously. Extensive experiments demonstrate the effectiveness of SM-HFEGCN in point cloud classification and segmentation tasks.

Combining multi-level feature extraction algorithm with residual graph convolutional neural network for partial discharge detection

Partial discharge (PD) identification is critical for the insulation diagnosis of cable conductors; however, there is still scope for enhancement in the existing adaptive extraction capabilities and feature utilization for PD signals. In this context, this paper introduces a method that integrates a one-dimensional convolutional neural network (1D-CNN) with a residual graph convolutional neural network (ResGCN) to recognize PD signals. Noise analysis is performed using various combinations of mother wavelets, and Bayesian optimization is employed to mitigate background noise. Key features of PD signals are progressively abstracted through a 1D-CNN-based multilevel automatic feature learning method, while signal timing attributes are maintained to minimize manual intervention. The graph data is constructed using the signal feature matrix and the signal timing feature similarity matrix. This is followed by the development of a ResGCN utilizing a graph attention mechanism to integrate node feature information and the topology of the PD graph data. This approach aims to fully exploit the correlation between local regions of the feature space and the temporal numerical properties of the signals. Additionally, it jointly optimizes feature extraction and model classification to facilitate adaptive diagnosis. The method is validated with extensive real experimental data obtained from medium voltage overhead power lines. It demonstrates exceptional performance and practicality, achieving an accuracy rate of 97.3% and a recognition rate of 96.1% for PD samples, thus offering reliable theoretical support for effective PD detection.

ACGRHA-Net: Accelerated Multi-Contrast MR Imaging with Adjacency Complementary Graph assisted Residual Hybrid Attention Network

Multi-contrast magnetic resonance (MR) imaging is an advanced technology used in medical diagnosis, but the long acquisition process can lead to patient discomfort and limit its broader application. Shortening acquisition time by undersampling k-space data introduces noticeable aliasing artifacts. To address this, we propose a method that reconstructs multi-contrast MR images from zero-filled data by utilizing a fully-sampled auxiliary contrast MR image as a prior to learn an adjacency complementary graph. This graph is then combined with a residual hybrid attention network, forming the adjacency complementary graph assisted residual hybrid attention network (ACGRHA-Net) for multi-contrast MR image reconstruction. Specifically, the optimal structural similarity is represented by a graph learned from the fully sampled auxiliary image, where the node features and adjacency matrices are designed to precisely capture structural information among different contrast images. This structural similarity enables effective fusion with the target image, improving the detail reconstruction. Additionally, a residual hybrid attention module is designed in parallel with the graph convolution network, allowing it to effectively capture key features and adaptively emphasize these important features in target contrast MR images. This strategy prioritizes crucial information while preserving shallow features, thereby achieving comprehensive feature fusion at deeper levels to enhance multi-contrast MR image reconstruction. Extensive experiments on the different datasets, using various sampling patterns and accelerated factors demonstrate that the proposed method outperforms the current state-of-the-art reconstruction methods.

Labeling small-degree nodes promotes semi-supervised community detection on graph convolutional network

Labeling small-degree nodes promotes semi-supervised community detection on graph convolutional network

Boosting multi-document summarization with hierarchical graph convolutional networks

The input of the multi-document summarization task is usually long and has high redundancy. Encoding multiple documents is a challenge for the Seq2Seq architecture. The way of concatenating multiple documents into a sequence ignores the relation between documents. Attention-based Seq2Seq architectures have slightly improved the cross-document relation modeling for multi-document summarization. However, these methods ignore the relation between sentences, and there is little improvement that can be achieved through the attention mechanism alone. This paper proposes a hierarchical approach to leveraging the relation between words, sentences, and documents for abstractive multi-document summarization. Our model employs the Graph Convolutional Networks (GCN) for capturing the cross-document and cross-sentence relations. The GCN module can enrich semantic representations by generating high-level hidden features. Our model achieves significant improvement over the attention-based baseline, beating the Hierarchical Transformer by 3.4/1.64, 1.92/1.44 ROUGE-1/2 F1 points on the Multi-News and WikiSum datasets, respectively. Experimental results demonstrate that our delivered method brings substantial improvements over several strong baselines.

Multi-scale structure-guided graph generation for multi-view semi-supervised classification

Graph convolutional network has emerged as a focal point in machine learning because of its robust graph processing capability. Most existing graph convolutional network-based approaches are designed for single-view data, yet in many practical scenarios, data is represented through multiple views. Moreover, due to the complexity of multiple views, normal graph generation methods can not mitigate redundancy to generate a high quality graph. Although the ability of graph convolutional network is undeniable, the quality of graph directly affects its performance. To tackle the aforementioned challenges, this paper proposes a multi-scale graph generation deep learning framework, called multi-scale semi-supervised graph generation based multi-view classification, consisting of two modules: edge sampling and path sampling. The former aims to generate an adjacency graph by selecting edges based on the maximum likelihood among graphs from different views. Meanwhile, the latter seeks to construct a adjacency graph according to the characteristics of paths within the graphs. Finally, the statistical technique is employed to extract commonality and generate a fused graph. Extensive experimental results robustly demonstrate the superior performance of our proposed framework, compared to other state-of-the-art multi-view semi-supervised approaches.

Graph-enhanced visual representations and question-guided dual attention for visual question answering

Visual Question Answering (VQA) has witnessed significant advancements recently, due to the application of deep learning in the field of vision-language research. Most current VQA models focus on merging visual and text features, but it is essential for these models to also consider the relationships between different parts of an image and use question information to highlight important features. This study proposes a method to enhance neighboring image region features and learn question-aware visual representations. First, we construct a region graph to represent spatial relationships between objects in the image. Then, graph convolutional networks (GCNs) are used to propagate information across neighboring regions, enriching each region’s feature representation by integrating contextual information. To capture long-range dependencies, the graph is enhanced with random walk with restart (RWR), enabling multi-hop reasoning across distant regions. Furthermore, a question-aware dual attention mechanism is introduced to further refine region features at both region and feature levels, ensuring that the model emphasizes key regions that are critical for answering the question. The enhanced region representations are then combined with the encoded question to predict an answer. Through extensive experiments on VQA benchmarks, the study demonstrates state-of-the-art performance by leveraging regional dependencies and question guidance. The integration of GCNs and random walks in the graph helps capture contextual information to focus visual attention selectively, resulting in significant improvements over existing methods on VQA 1.0 and VQA 2.0 benchmark datasets.

GACT-PPIS: Prediction of protein-protein interaction sites based on graph structure and transformer network

The prediction of protein-protein interaction sites (PPIS) is currently crucial for regulating many biological activities in cells and developing drugs for various diseases. Deep learning-based methods have been proposed for predicting PPIS, significantly reducing the manpower and time costs associated with traditional experimental methods such as yeast two-hybrid, mass spectrometry, and affinity purification. However, the predictive accuracy of these deep learning methods has not yet reached the expected level. Therefore, we introduce a model called GACT-PPIS.The design of the GACT-PPIS algorithm aims to utilize combined information from protein sequences and structures as input to predict protein-protein interaction sites. The core of GACT-PPIS utilizes an Enhanced Graph Attention Network (EGAT) with initial residual and identity mappings, along with a deep Transformer network as the basic units, supplemented by Graph Convolutional Networks (GCN), effectively aggregating information from neighboring nodes for each node. After multiple network layers, the information of the entire protein is also fused into the nodes, and the Transformer network further enhances the model's performance.Experimental results show that GACT-PPIS outperforms the most representative models in terms of Recall, F1-measure, MCC, AUROC, and AUPRC on the benchmark test set (Test-60). Additionally, on other independent test sets (UBTest-31-6), GACT-PPIS leads in terms of Accuracy, Precision, Recall, F1-measure, MCC, AUROC, and AUPRC compared to the most representative models. It is worth noting that GACT-PPIS demonstrates excellent generalization and versatility across different test sets, showcasing good performance on multiple test sets for the same trained GACT-PPIS model.

Intelligible graph contrastive learning with attention-aware for recommendation

Recommender systems are an important tool for information retrieval, which can aid in the solution of the issue of information overload. Recently, contrastive learning has shown remarkable performance in recommendation by data augmentation processes to address highly sparse data. Our paper proposes an Intelligible Graph Contrastive Learning with attention-aware (IntGCL) for recommendation. Particularly, our IntGCL first introduces a novel attention-aware matrix into graph convolutional networks (GCN) to identify the importance between users and items, which is constructed to preserve the importance between users and items by a random walk with a restart strategy and can enhance the intelligibility of our model. Then, the attention-aware matrix is further utilised to guide the generation of a graph-generative model with attention-aware and a graph-denoising model for automatically generating two trainable contrastive views for data augmentation, which can de-noise and further enhance the intelligibility. Comprehensive experiments on four real-world datasets indicate the superiority of our IntGCL approach over multiple state-of-the-art methods. Our datasets and source code are available at https://github.com/restarthxr/InpGCL.

Integrating hybrid deep learning and path allocation for real-time inbound passenger flow prediction and anomaly detection in urban rail transit

This paper study the problem of real-time prediction of inbound passenger flow and the detection and alerting of abnormal passenger flows in urban rail transit (URT) networks. We propose a fused framework that combines a hybrid deep learning model and an evaluation strategy. Specifically, the learning model incorporates Graph Convolutional Networks (GCN), Gated Recurrent Units (GRU), and attention mechanisms to effectively capture spatial and temporal correlations in passenger flow data. The evaluation strategy utilizes a depth-first search algorithm to determine the optimal travel paths for each individual passenger. And based on the paths, we develop a real-time method for estimating the origin–destination (OD) matrix that utilizes both long-term and short-term historical destination trend vectors to reduce dimensions while improving predictive accuracy. Through extensive testing using data from the Shanghai rail transit system, we demonstrate that this fused framework achieves high prediction accuracy for inbound passenger flow at various stations while efficiently identifying and warning sudden large-scale events involving significant increases in passenger flow volume. This research contributes towards improving overall passenger experience as well as operational resilience within urban rail systems when dealing with large-scale influxes of passengers.

HOIMotion: Forecasting Human Motion During Human-Object Interactions Using Egocentric 3D Object Bounding Boxes.

We present HOIMotion - a novel approach for human motion forecasting during human-object interactions that integrates information about past body poses and egocentric 3D object bounding boxes. Human motion forecasting is important in many augmented reality applications but most existing methods have only used past body poses to predict future motion. HOIMotion first uses an encoder-residual graph convolutional network (GCN) and multi-layer perceptrons to extract features from body poses and egocentric 3D object bounding boxes, respectively. Our method then fuses pose and object features into a novel pose-object graph and uses a residual-decoder GCN to forecast future body motion. We extensively evaluate our method on the Aria digital twin (ADT) and MoGaze datasets and show that HOIMotion consistently outperforms state-of-the-art methods by a large margin of up to 8.7% on ADT and 7.2% on MoGaze in terms of mean per joint position error. Complementing these evaluations, we report a human study (N=20) that shows that the improvements achieved by our method result in forecasted poses being perceived as both more precise and more realistic than those of existing methods. Taken together, these results reveal the significant information content available in egocentric 3D object bounding boxes for human motion forecasting and the effectiveness of our method in exploiting this information.

Deep Reinforcement Learning of Graph Convolutional Neural Network for Resilient Production Control of Mass Individualized Prototyping Toward Industry 5.0

Deep Reinforcement Learning of Graph Convolutional Neural Network for Resilient Production Control of Mass Individualized Prototyping Toward Industry 5.0

Automatic 3-dimensional quantification of orthodontically induced root resorption in cone-beam computed tomography images based on deep learning

Orthodontically induced root resorption (OIRR) is a common and undesirable consequence of orthodontic treatment. Traditionally, studies employ manual methods to conduct 3-dimensional quantitative analysis of OIRR via cone-beam computed tomography (CBCT), which is often subjective and time-consuming. With advancements in computer technology, deep learning-based approaches have gained traction in medical image processing. This study presents a deep learning-based model for the fully automatic extraction of root volume information and the localization of root resorption from CBCT images. In this cross-sectional, retrospective study, 4534 teeth from 105 patients were used to train and validate an automatic model for OIRR quantification. The protocol encompassed several steps: preprocessing of CBCT images involving automatic tooth segmentation and conversion into point clouds, followed by segmentation of tooth crowns and roots via the Dynamic Graph Convolutional Neural Network. The root volume was subsequently calculated, and OIRR localization was performed. The intraclass correlation coefficient was employed to validate the consistency between the automatic model and manual measurements. The proposed method strongly correlated with manual measurements in terms of root volume and OIRR severity assessment. The intraclass correlation coefficient values for average volume measurements at each tooth position exceeded 0.95 (P<0.001), with the accuracy of different OIRR severity classifications surpassing 0.8. The proposed methodology provides automatic and reliable tools for OIRR assessment, offering potential improvements in orthodontic treatment planning and monitoring.

A histogram-based approach to calculate graph similarity using graph neural networks

Deep learning has revolutionized the field of pattern recognition and machine learning by exhibiting exceptional efficiency in recognizing patterns. The success of deep learning can be seen in a wide range of applications including speech recognition, natural language processing, video processing, and image classification. Moreover, it has also been successful in recognizing structural patterns, such as graphs. Graph Neural Networks (GNNs) are models that employ message passing between nodes in a graph to capture its dependencies. These networks memorize a state that approximates graph information with greater depth compared to traditional neural networks. Although training a GNN can be challenging, recent advances in GNN variants, including Graph Convolutional Neural Networks, Gated Graph Neural Networks, and Graph Attention Networks, have shown promising results in solving various problems. In this work, we present a GNN-based approach for computing graph similarity and demonstrate its application to a classification problem. Our proposed method converts the similarity of two graphs into a score, and experiments on state-of-the-art datasets show that the proposed technique is effective and efficient. Results are summarized using a confusion matrix and mean square error metric, demonstrating the accuracy of our proposed technique.