Graph Data Research Articles

Domination based graph neural networks

Graph Neural Networks (GNNs) have emerged as a widely used and effective method across various domains for learning from graph data. Despite the abundance of GNN variants, many struggle with effectively propagating messages over long distances. This paper introduces a novel hierarchical message passing framework for graph learning, specifically designed to address the challenge of long-distance message propagation in graphs. By constructing smaller graphs from the main graph using the concept of domination, a fundamental principle in graph theory, we facilitate more efficient message passing within each subgraph. Subsequently, we employ a Graph Attention Network (GAT) to aggregate these features and propagate them to distant nodes across the graph. Experimental results on standard node classification datasets validate that the proposed architecture achieves performance comparable to or better than conventional GNNs. Additionally, our model consistently performs better on graphs with missing edges.

SPMGAE: Self-purified masked graph autoencoders release robust expression power

To tackle the scarcity of labeled graph data, graph self-supervised learning (SSL) has branched into two paradigms: Generative methods and Contrastive methods. Inspired by MAE and BERT in computer vision (CV) and natural language processing (NLP), masked graph autoencoders (MGAEs) are gaining popularity in the generative genre. However, prevailing MGAEs are mostly designed under the assumption that the data has high homophilic score and is out of adversarial distortion. When people deliberately improve the performance on homophilic graph datasets, they ignore a critical issue that both internal heterophily and artificial attack noise are quite common in the real world. Therefore, when data itself is highly heterophilic or confronted with attacks, they merely have no defensive capability. Especially under self-supervised conditions, it is much more difficult to detect internal heterophily and resist artificial attacks. In this paper, we propose a Self-Purified Masked Graph Autoencoder (SPMGAE) to make up for the shortcomings of prevailing MGAEs in terms of robustness. SPMGAE first utilizes a self-purified module to prune raw graph data and separate perturbation information. The purified graph provides a robust graph structure for the entire pre-training process. Next, the encoding module reuses perturbation information for auxiliary training to enhance robustness, while the decoding module reconstructs the effective graph data at a finer granularity. Extensive experiments on homophilic and heterophilic datasets attacked by various attack methods demonstrate SPMGAE has a considerable robust expressive ability. Especially on small datasets with large perturbations, the improvement of defensive performance could reaches 10%–25%.

A review of feature selection strategies utilizing graph data structures and Knowledge Graphs.

Feature selection in Knowledge Graphs (KGs) is increasingly utilized in diverse domains, including biomedical research, Natural Language Processing (NLP), and personalized recommendation systems. This paper delves into the methodologies for feature selection (FS) within KGs, emphasizing their roles in enhancing machine learning (ML) model efficacy, hypothesis generation, and interpretability. Through this comprehensive review, we aim to catalyze further innovation in FS for KGs, paving the way for more insightful, efficient, and interpretable analytical models across various domains. Our exploration reveals the critical importance of scalability, accuracy, and interpretability in FS techniques, advocating for the integration of domain knowledge to refine the selection process. We highlight the burgeoning potential of multi-objective optimization and interdisciplinary collaboration in advancing KG FS, underscoring the transformative impact of such methodologies on precision medicine, among other fields. The paper concludes by charting future directions, including the development of scalable, dynamic FS algorithms and the integration of explainable AI principles to foster transparency and trust in KG-driven models.

On the Inference of Original Graph Information from Graph Embeddings

Graph embedding converts a graph data into a low dimensional space to preserve the original graph information. However, graph data can be reconstructed by malicious adversaries to train machine learning models from graph embeddings. This paper studies to what extent an adversary (without the original graph data) can recover the original graph data from graph embeddings. To quantify the original graph information leakage from graph embeddings, we develop a deep neural network model InferNet that can be used by adversaries to infer the original graph information from an adversary-accessible graph embedding database. More specifically, we propose the data-free reversed knowledge distillation technique to support the InferNet training even if the original graph dataset is absent. To ensure the performance of InferNet, we design two cycle-consistency loss functions to have an interactive training of InferNet over three series of datasets. To further enhance the performance of InferNet, we provide a joint training algorithm that simultaneously trains the pseudo-sample generator and InferNet, which significantly reduces the storage space. We evaluate the performance of InferNet on three datasets, and the intensive experiments demonstrate that InferNet can infer the original graph information from the graph embedding dataset with high accuracy.

Asymmetric augmented paradigm-based graph neural architecture search

In most scenarios of graph-based tasks, graph neural networks (GNNs) are trained end-to-end with labeled samples. Labeling graph samples, a time-consuming and expert-dependent process, leads to huge costs. Graph data augmentations can provide a promising method to expand labeled samples cheaply. However, graph data augmentations will damage the capacity of GNNs to distinguish non-isomorphic graphs during the supervised graph representation learning process. How to utilize graph data augmentations to expand labeled samples while preserving the capacity of GNNs to distinguish non-isomorphic graphs is a challenging research problem. To address the above problem, we abstract a novel asymmetric augmented paradigm in this paper and theoretically prove that it offers a principled approach. The asymmetric augmented paradigm can preserve the capacity of GNNs to distinguish non-isomorphic graphs while utilizing augmented labeled samples to improve the generalization capacity of GNNs. To be specific, the asymmetric augmented paradigm will utilize similar yet distinct asymmetric weights to classify the real sample and augmented sample, respectively. To systemically explore the benefits of asymmetric augmented paradigm under different GNN architectures, rather than studying individual asymmetric augmented GNN (A2GNN) instance, we then develop an auto-search engine called Asymmetric Augmented Graph Neural Architecture Search (A2GNAS) to save human efforts. We empirically validate our asymmetric augmented paradigm on multiple graph classification benchmarks, and demonstrate that representative A2GNN instances automatically discovered by our A2GNAS method achieve state-of-the-art performance compared with competitive baselines. Our codes are available at: https://github.com/csubigdata-Organization/A2GNAS.

Thermoelastic damping in asymmetric vibrations of nonlocal circular plate resonators with Moore-Gibson-Thompson heat conduction

Precise estimation of the amount of thermoelastic damping (TED) in small-sized resonators is one of the crucial issues in their optimal design. According to several experimental, numerical and theoretical findings, the behavior of structures with such small dimensions in structural and thermal domains does not follow classical elasticity theory and the Fourier law of heat conduction. The objective of this work is to develop a nonclassical formulation for computing TED in asymmetric vibrations of circular nanoplates according to nonlocal theory (NT) and Moore-Gibson-Thompson (MGT) heat transfer model. To do so, first, the coupled equations of motion and heat are derived in the purview of NT and MGT model. Then, the real and imaginary parts of the system frequency are extracted from the frequency equation affected by thermoelastic coupling. Finally, by means of the existing definition for TED, a mathematical expression including nonlocal parameter of NT and nonclassical constants of MGT model is obtained to make an estimate of TED value. In the numerical examples section, to ensure the credibility of the provided formulation, the results obtained through this model are compared with those reported in the literature in the context of simpler models. Then, by presenting various graphical data, the type and extent of the role of different factors in TED changes are appraised. The obtained results indicate that employing NT and MGT model can substantially impact the amount of TED compared to the estimations of the classical formulation. Furthermore, the comparison of results reveals that the magnitude of TED in asymmetric vibration modes surpasses that in symmetric ones, and consequently incorporating asymmetric vibration modes is crucial for thorough analysis of TED in circular nanoplates. Thus, the formulation presented in this article can aid in achieving an optimal design for such plates, particularly in terms of minimizing energy loss.

Portable Network Resolving Huge-Graph Isomorphism Problem

Abstract The graph isomorphism, as a key task in graph data analysis, is of great significance for the understanding, feature extraction, and pattern recognition of graph data. The best performance of traditional methods is the quasi-polynomial time complexity, which is infeasible for huge graphs. This paper aims to propose a lightweight graph network to resolve the problem of isomorphism in huge graphs. We propose a partitioning algorithm with linear time complexity based on isomorphic necessary condition to handle network graphs that cannot be fully computed by a single computer. We use anti-weight convolution analysis and skip connections to obtain a more stable representation with fewer layers and parameters. We implement simulations on personal computer (PC) with graphs consisting of hundred millions of edges, demonstrating the identification performance ($>98\%$) and time efficiency.

Analysis of counselor task commitment as a professional characteristic in junior and senior high schools in West Sumatra

Task commitment is an important thing for counselors to have so that the counseling profession provides professional services to various stakeholders. This research aims to look at the job commitment profile of middle school and high school level counselors so that it can be used as an evaluation regarding the counselor's current performance or services. The population in this study were counselors in eight cities in West Sumatra. This study uses a quantitative approach. The research sample was determined using an accidental sampling technique (convenience sampling), namely the sampling process simply involves taking the counselors the observer meets in the field. The research sample obtained was 86 counselors. The data collection technique was carried out by distributing questionnaires. The collected data was analyzed quantitatively using the SPSS 25 for Windows program and the Excel application for processing quantitative descriptive analysis, categorization and percentages and producing graphical data. The research results show that the majority of counselors' task commitment is in the very high category for all indicators, namely tough attitude, tenacity, not getting bored easily, independence, courage to take risks, desire to improve themselves, and desire to succeed.

Vertex-based graph neural network classification model considering structural topological features for structural optimization

Traditional surrogate models always face the challenge of low accuracy when dealing with high-dimensional problems in structural optimization, this study aims to overcome this problem and proposes a vertex-based graph neural network (GNN) classification model. In contrast to conventional machine learning models that treat design variables as independent inputs, the proposed model develops a vertex-based graph representation to transform structural topological features and critical physical information into the graph data. According to a message passing mechanism based on the graph convolutional, it can extract the correlations among design variables and enhance its capability in handling high-dimensional structural optimization problems. Three truss examples, including a 10-bar with 10 variables, a 600-bar with 25 variables, and a 942-bar with 59 variables, are utilized to investigate the performance of the proposed surrogate model. The results demonstrate that the GNN-based surrogate model outperforms traditional machine learning approaches, particularly in the two high-dimensional problems, showcasing its superior ability to capture complex variable correlations and handle high-dimensional structural optimization tasks. Moreover, the proposed method significantly reduces the computational expenses by over 60% compared to conventional metaheuristic algorithms, while yielding optimal designs with comparable quality. These results demonstrate the efficiency and effectiveness of the GNN-based surrogate model in tackling complex, high-dimensional structural optimization problems.

Visual analysis of multi-omics data.

We present a tool for multi-omics data analysis that enables simultaneous visualization of up to four types of omics data on organism-scale metabolic network diagrams. The tool's interactive web-based metabolic charts depict the metabolic reactions, pathways, and metabolites of a single organism as described in a metabolic pathway database for that organism; the charts are constructed using automated graphical layout algorithms. The multi-omics visualization facility paints each individual omics dataset onto a different "visual channel" of the metabolic-network diagram. For example, a transcriptomics dataset might be displayed by coloring the reaction arrows within the metabolic chart, while a companion proteomics dataset is displayed as reaction arrow thicknesses, and a complementary metabolomics dataset is displayed as metabolite node colors. Once the network diagrams are painted with omics data, semantic zooming provides more details within the diagram as the user zooms in. Datasets containing multiple time points can be displayed in an animated fashion. The tool will also graph data values for individual reactions or metabolites designated by the user. The user can interactively adjust the mapping from data value ranges to the displayed colors and thicknesses to provide more informative diagrams.

Graph neural network coarse-grain force field for the molecular crystal RDX

Condense phase molecular systems organize in wide range of distinct molecular configurations, including amorphous melt and glass as well as crystals often exhibiting polymorphism, that originate from their intricate intra- and intermolecular forces. While accurate coarse-grain (CG) models for these materials are critical to understand phenomena beyond the reach of all-atom simulations, current models cannot capture the diversity of molecular structures. We introduce a generally applicable approach to develop CG force fields for molecular crystals combining graph neural networks (GNN) and data from an all-atom simulations and apply it to the high-energy density material RDX. We address the challenge of expanding the training data with relevant configurations via an iterative procedure that performs CG molecular dynamics of processes of interest and reconstructs the atomistic configurations using a pre-trained neural network decoder. The multi-site CG model uses a GNN architecture constructed to satisfy translational invariance and rotational covariance for forces. The resulting model captures both crystalline and amorphous states for a wide range of temperatures and densities.

Fiber laser-based two-wavelength sensors for detecting temperature and strain on concrete structures

In this article, we discuss the scientific research we conducted to ensure information security on concrete structures using the fiber Bragg grating method. As a result, guided by the experiences of other researchers and their results in our studies, the main hazards were identified, and the influence of temperature and deformation sensors in concrete was studied. The work determines the dependence of temperature on the optical cable and its effect on concrete deformation. It also studies the change in the internal structure of the optical fiber when exposed to concrete, and in comparison, with non-concrete optical fiber. In the experimental chapter, conducted on laboratory bench, we consider and study the signal transmission structure of an optical sensor that experiences deformation upon impact with concrete and such fibers. Graphic data was obtained using the OptiSystem program. During the process of determining the optical power of the sensors, we discovered that the higher the vibration during signal transmission, the higher the power of the optical data acquisition system unit. The results presented in this paper show the promise of fiber optics in other environments, as well as new discoveries in the construction industry and the successful use of electrical laser sensors to maintain temperature and stress in a variety of concrete structures, including cooling towers in nuclear and thermal power plants.

Automating Government Report Generation: A Generative AI Approach for Efficient Data Extraction, Analysis, and Visualization

This application paper introduces a transformative solution to address the labour-intensive manual report generation, data searching & report revision process in government entities. Traditional methods of data extraction, analysis, and graph creation for annual reports are not only time-consuming but also prone to human errors. To mitigate these challenges, we propose an innovative system leveraging Generative Artificial Intelligence (GenAI), with a specific focus on large language models (LLMs). Our solution incorporates automated data extraction from diverse sources designated as internal knowledge base, text analysis, summarization using advanced language models, and the generation as well as revisions of informative graphs. Different LLMs like Google's Gemini Pro and OpenAI's GPT 4.0 has been used to read different data visualisation graphs and fetching the information from internal knowledge base, respectively, to update the graphs in automated manner. Solution implementation using Python language on the World Economic Situation and Prospects report by Department of Economic & Social Affairs, United Nation shows that the early result produces almost 0.87% to 17.46% average error rates in the task of factual data visualisation graph. Key benefit of this approach include improved time efficiency, consistency in report format, enhanced insights, and a user-friendly interface. A review and approval workflow facilitate user feedback, contributing to continuous model performance improvement.

Finding Efficient Graph Embeddings and Processing them by a CNN-based Tool

We introduce new tools to support finding efficient graph embedding techniques for graph databases and to process their outputs using deep learning for classification scenarios. Accordingly, we investigate the possibility of creating an ensemble of different graph embedding methods to raise accuracy and present an interconnected neural network-based ensemble to increase the efficiency of the member classification algorithms. We also introduce a new convolutional neural network-based architecture that can be generally proposed to process vectorized graph data provided by various graph embedding methods and compare it with other architectures in the literature to show the competitiveness of our approach. We also exhibit a statistical-based inhomogeneity level estimation procedure to select the optimal embedding for a given graph database efficiently. The efficiency of our framework is exhaustively tested using several publicly available graph datasets and numerous state-of-the-art graph embedding techniques. Our experimental results for classification tasks have proved the competitiveness of our approach by outperforming the state-of-the-art frameworks.

Affinity Uncertainty-Based Hard Negative Mining in Graph Contrastive Learning.

Hard negative mining has shown effective in enhancing self-supervised contrastive learning (CL) on diverse data types, including graph CL (GCL). The existing hardness-aware CL methods typically treat negative instances that are most similar to the anchor instance as hard negatives, which helps improve the CL performance, especially on image data. However, this approach often fails to identify the hard negatives but leads to many false negatives on graph data. This is mainly due to that the learned graph representations are not sufficiently discriminative due to oversmooth representations and/or non-independent and identically distributed (non-i.i.d.) issues in graph data. To tackle this problem, this article proposes a novel approach that builds a discriminative model on collective affinity information (i.e., two sets of pairwise affinities between the negative instances and the anchor instance) to mine hard negatives in GCL. In particular, the proposed approach evaluates how confident/uncertain the discriminative model is about the affinity of each negative instance to an anchor instance to determine its hardness weight relative to the anchor instance. This uncertainty information is then incorporated into the existing GCL loss functions via a weighting term to enhance their performance. The enhanced GCL is theoretically grounded that the resulting GCL loss is equivalent to a triplet loss with an adaptive margin being exponentially proportional to the learned uncertainty of each negative instance. Extensive experiments on ten graph datasets show that our approach does the following: 1) consistently enhances different state-of-the-art (SOTA) GCL methods in both graph and node classification tasks and 2) significantly improves their robustness against adversarial attacks. Code is available at https://github.com/mala-lab/AUGCL.

Multi-Domain Based Dynamic Graph Representation Learning for EEG Emotion Recognition.

Graph neural networks (GNNs) have demonstrated efficient processing of graph-structured data, making them a promising method for electroencephalogram (EEG) emotion recognition. However, due to dynamic functional connectivity and nonlinear relationships between brain regions, representing EEG as graph data remains a great challenge. To solve this problem, we proposed a multi-domain based graph representation learning (MD 2GRL) framework to model EEG signals as graph data. Specifically, MD 2GRL leverages gated recurrent units (GRU) and power spectral density (PSD) to construct node features of two subgraphs. Subsequently, the self-attention mechanism is adopted to learn the similarity matrix between nodes and fuse it with the intrinsic spatial matrix of EEG to compute the corresponding adjacency matrix. In addition, we introduced a learnable soft thresholding operator to sparsify the adjacency matrix to reduce noise in the graph structure. In the downstream task, we designed a dual-branch GNN and incorporated spatial asymmetry for graph coarsening. We conducted experiments using the publicly available datasets SEED and DEAP, separately for subject-dependent and subject-independent, to evaluate the performance of our model in emotion classification. Experimental results demonstrated that our method achieved state-of-the-art (SOTA) classification performance in both subject-dependent and subject-independent experiments. Furthermore, the visualization analysis of the learned graph structure reveals EEG channel connections that are significantly related to emotion and suppress irrelevant noise. These findings are consistent with established neuroscience research and demonstrate the potential of our approach in comprehending the neural underpinnings of emotion.

Abstract P348: Effects of Different Antihypertensive Regimens on Cognitive Function and Blood Pressure Changes in Elderly Hypertensive Patients: Network Meta-Analysis

Background: Elderly hypertensive patients with cognitive impairment not only exhibit reduced processing speed and executive function but also suffer significant deficits in memory and motor speed, adversely affecting their quality of life. This study employs a network meta-analysis to evaluate the effects of different antihypertensive regimens on cognitive function and blood pressure changes in elderly hypertensive patients. Methods: A comprehensive search was conducted in PubMed, Cochrane Library, Embase, Web of Science, CNKI, Wanfang and VIP databases for randomized controlled trials on different antihypertensive regimens for elderly hypertensive patients. Data analysis and graphing were performed using Review Manager 5.3 and Stata 16.0 software. Results: A total of 16 studies involving 11,694 elderly hypertensive patients and six antihypertensive regimens were included. The network meta-analysis results indicated that elderly hypertensive patients treated with angiotensin II receptor blockers (ARBs) had a greater change in cognitive function scores compared to those receiving usual care (standardized mean difference [SMD] = 0.46, 95% CI [0.23, 0.70]), angiotensin-converting enzyme inhibitors (ACEIs) (SMD = 0.33, 95% CI [0.06, 0.61]), beta-blockers (BBs) (SMD = 0.48, 95% CI [0.17, 0.78]), and diuretics (SMD = 0.32, 95% CI [0.01, 0.63]). Similarly, patients treated with calcium channel blockers (CCBs) showed greater changes in cognitive function scores compared to those receiving usual care (SMD = 0.37, 95% CI [0.04, 0.70]) and BBs (SMD = 0.38, 95% CI [0.02, 0.74]). The SUCRA rankings for changes in cognitive function scores among different antihypertensive regimens were ARB (93.6%), CCB (81.2%), diuretics (47.1%), ACEI (43.4%), usual care (17.9%), and BB (16.8%). Conclusion: The current evidence suggests that, compared to usual care, ACEIs, BBs, CCBs, and diuretics, ARBs might be the optimal regimen for improving cognitive function in elderly hypertensive patients.

Memristive Crossbar Array-Based Probabilistic Graph Modeling.

Modern graph datasets with structural complexity and uncertainties due to incomplete information or data variability require advanced modeling techniques beyond conventional graph models. This study introduces a memristive crossbar array (CBA)-based probabilistic graph model (C-PGM) utilizing Cu0.3Te0.7/HfO2/Pt memristors, which exhibit probabilistic switching, self-rectifying, and memory characteristics. C-PGM addresses the complexities and uncertainties inherent in structural graph data across various domains, leveraging the probabilistic nature of memristors. C-PGM relies on the device-to-device variation across multiple memristive CBAs, overcoming the limitations of previous approaches that rely on sequential operations, which are slower and have a reliability concern due to repeated switching. This new approach enables the fast processing and massive implementation of probabilistic units at the expense of chip area. In this study, the hardware-based C-PGM feasibly expresses small-scale probabilistic graphs and shows minimal error in aggregate probability calculations. The probability calculation capabilities of C-PGM are applied to steady-state estimation and the PageRank algorithm, which is implemented on a simulated large-scale C-PGM. The C-PGM-based steady-state estimation and PageRank algorithm demonstrate comparable accuracy to conventional methods while significantly reducing computational costs.

Electric vehicle charging load prediction based on graph attention networks and autoformer

AbstractWith the widespread popularity of electric vehicles in the domestic market, large‐scale electric vehicle user data has been collected and stored. Highly accurate user‐level charging load prediction has a wide range of application scenarios and great business value. However, most existing EV load prediction methods are modelled from the charging station perspective, ignoring the user's travel habits and charging demand. Therefore, this paper proposes a temporal spatial neural network based on graph attention and Autoformer to predict electric vehicle charging load. Firstly, the urban map of Wuhan is rasterized. Then, driving and charging data from the user level are aggregated into the raster module according to the time sequence, and a spatio‐temporal graph data structure of user travel trajectory is constructed. Finally, the temporal spatial neural network is used to construct the EV charging load prediction model from the user's perspective. The experimental results show that, compared with other baseline prediction methods, the proposed method effectively improves the accuracy of the EV charging load prediction model by fully exploiting the distribution of EV user clusters in time and geographic space.

Toward Robust Graph Semi-Supervised Learning Against Extreme Data Scarcity.

The success of graph neural networks (GNNs) in graph-based web mining highly relies on abundant human-annotated data, which is laborious to obtain in practice. When only a few labeled nodes are available, how to improve their robustness is key to achieving replicable and sustainable graph semi-supervised learning. Though self-training is powerful for semi-supervised learning, its application on graph-structured data may fail because 1) larger receptive fields are not leveraged to capture long-range node interactions, which exacerbates the difficulty of propagating feature-label patterns from labeled nodes to unlabeled nodes and 2) limited labeled data makes it challenging to learn well-separated decision boundaries for different node classes without explicitly capturing the underlying semantic structure. To address the challenges of capturing informative structural and semantic knowledge, we propose a new graph data augmentation framework, augmented graph self-training (AGST), which is built with two new (i.e., structural and semantic) augmentation modules on top of a decoupled GST backbone. In this work, we investigate whether this novel framework can learn a robust graph predictive model under the low-data context. We conduct comprehensive evaluations on semi-supervised node classification under different scenarios of limited labeled-node data. The experimental results demonstrate the unique contributions of the novel data augmentation framework for node classification with few labeled data.