Graph Data Research Articles

Dynamic graph structure and spatio-temporal representations in wind power forecasting

Wind Power Forecasting (WPF) has gained considerable focus as a crucial aspect of the successful integration and operation of wind power. However, due to the stochastic and unstable nature of wind, it poses a real challenge to effectively analyze the correlations among multiple time series data for accurate prediction. In our study, an end-to-end framework called Dynamic Graph structure and Spatio-Temporal representation learning (DSTG) framework is proposed to achieve stable power forecasting by constructing graph data to capture the critical features in the data. Specifically, a graph structure learning (GSL) module is introduced to dynamically construct task-related correlation matrices via backpropagation to mitigate the inherent inconsistency and randomness of wind power data. Additionally, a dual-scale temporal graph learning (DTG) module is further proposed to explore the implicit spatio-temporal features at a fine-grained level using different skip connections from the constructed graph data. Finally, comprehensive experiments are performed on the collected Xuji Group Wind Power (XGWP) dataset, and the results show that DSTG outperforms the state-of-the-art spatio-temporal methods by 10.12% on the average of root mean square error and mean absolute error, demonstrating the effectiveness of DSTG. In conclusion, our model provides a promising approach.

Graph Neural Networks with Multi-features for Predicting Cocrystals using APIs and Coformers Interactions.

Active pharmaceutical ingredients (APIs) have gained direct pharmaceutical interest, along with their in vitro properties, and thus utilized as auxiliary solid dosage forms upon FDA guidance and approval on pharmaceutical cocrystals when reacting with coformers, as a potential and attractive route for drug substance development. However, screening and selecting suitable and appropriate coformers that may potentially react with APIs to successfully form cocrystals is a time-consuming, inefficient, economically expensive, and labour-intensive task. In this study, we implemented GNNs to predict the formation of cocrystals using our introduced API-coformers relational graph data. We further compared our work with previous studies that implemented descriptor-based models (e.g., random forest, support vector machine, extreme gradient boosting, and artificial neural networks). All built graph-based models show compelling performance accuracies (i.e., 91.36, 94.60 and 95. 95% for GCN, GraphSAGE, and RGCN respectively). RGCN demonstrated effectiveness and prevailed among the built graph-based models due to its capability to capture intricate and learn nuanced relationships between entities such as non-ionic and non-covalent interactions or link information between APIs and coformers which are crucial for accurate predictions and representations. These capabilities allows the model to adeptly learn the topological structure inherent in the graph data.

Geometric Morphometric Analysis of Plastinated Brain Sections Using Computer-Based Methods: Evaluating Shrinkage and Shape Changes.

Plastination preserves biological specimens for long-term and geometric morphometry analyzes shape differences with advanced statistical methods. This study primarily aimed to statistically quantify shrinkage in brain sections following plastination. The secondary goal was to present the shrinkage occurring in both tissues and cavities of specific anatomical structures using geometric morphometry. Fourteen sections from each of nine male bovine brains underwent silicone plastination using the standard technique, which involves four stages: fixation, dehydration, forced impregnation, and curing. The shrinkage percentage in plastinated sections was measured using ImageJ, while geometric morphometry was used for shape analysis. Correlation analysis was performed to reveal the relationship between ventricle area and shrinkage percentage. Significant shrinkage was observed in each brain section. A positive correlation was observed between the shrinkage on the sections and ventricular area. However, with the exception of section number 9, this correlation was not statistically significant for the remaining sections. Furthermore, the strength of the correlation exhibited variability across different sections. Using geometric morphometric, shrinkage, and associated shape variations in specific anatomical regions within the serial brain sections were illustrated with visual graphics. Shape analysis revealed the most pronounced shrinkage in the sulcus coronalis, sulcus suprasylvius rostralis, ventral surface of the thalamus, and the third ventricle. Additionally, the cornu ammonis (hippocampus) exhibited the most significant shrinkage and shape variation. While morphometric analyses did not reveal significant shrinkage in cavity-like structures, geometric morphometric analyses demonstrated significant shrinkage in the third ventricle. This study is unique in that it is the first to comprehensively demonstrate, using geometric morphometry, the morphological changes that occur in brain tissue during plastination. The findings, when combined with graphical and statistical data analysis, emphasize the effectiveness of geometric morphometry as a powerful tool for elucidating shape changes in plastination research.

Multisource representation learning for pediatric knowledge extraction from electronic health records.

Electronic Health Record (EHR) systems are particularly valuable in pediatrics due to high barriers in clinical studies, but pediatric EHR data often suffer from low content density. Existing EHR code embeddings tailored for the general patient population fail to address the unique needs of pediatric patients. To bridge this gap, we introduce a transfer learning approach, MUltisource Graph Synthesis (MUGS), aimed at accurate knowledge extraction and relation detection in pediatric contexts. MUGS integrates graphical data from both pediatric and general EHR systems, along with hierarchical medical ontologies, to create embeddings that adaptively capture both the homogeneity and heterogeneity between hospital systems. These embeddings enable refined EHR feature engineering and nuanced patient profiling, proving particularly effective in identifying pediatric patients similar to specific profiles, with a focus on pulmonary hypertension (PH). MUGS embeddings, resistant to negative transfer, outperform other benchmark methods in multiple applications, advancing evidence-based pediatric research.

MalenoCare - Skin Cancer Detection and Prescription using CNN and ML

Machine learning (ML) and convolutional neural networks (CNNs) have driven significant advancements in healthcare, particularly in dermatology, by enabling the automation of diagnostic processes for skin cancer. Skin cancer, being one of the most common types of cancer, requires early detection and accurate classification to improve patient outcomes and reduce mortality rates. This review paper explores various studies and CNN-based tools that enhance skin cancer detection and support preventive care through image analysis. The paper also discusses the effectiveness of different CNN architectures, including VGG16, ResNet, and Inception, in achieving high accuracy rates in skin lesion classification. A combination of recent research findings, model evaluation metrics, and graphical data highlights the accuracy, interpretability, and real-world applications of ML models, offering insights into their potential for integration into clinical practice.

A novel binary hashing for agricultural scenery classification.

In this research, we present PerceptHashing, a technique designed to categorize million-scale agricultural scenic images by incorporating human gaze shifting paths (GSPs) into a hashing framework. For each agricultural image, we identify visually and semantically significant object patches, such as fields, crops, and water bodies. These patches are linked to form a graphlet, establishing a network of spatially adjacent patches, and a GSP is then extracted using an active learning algorithm. The GSP reflects the distribution of human gaze across different regions of each agricultural scene, typically involving fewer than 12 regions, as validated by cross-validation. We then design a binary hashing framework that effectively leverages the semantics encoded in these GSPs. This framework integrates three key elements: (i) refinement of noisy labels, (ii) incorporation of deep image-level agricultural semantics, and (iii) updates to the adaptive data graph. The resulting hash codes from each GSP are converted into a kernelized visual descriptor for classification. To evaluate the influence of GSPs on agricultural image classification both qualitatively and quantitatively, we conducted an extensive user study comparing GSPs from typical observers with those from Alzheimer's patients. The results demonstrate that: (1) the classification accuracy of 1.22 million agricultural aerial images using our approach is significantly higher than those processed by other methods, and (2) GSPs from 33 Alzheimer's patients differ markedly from those of 37 normal observers, leading to notable differences in classification accuracy ( versus ).

DGNMDA: Dual Heterogeneous Graph Neural Network Encoder for miRNA-Disease Association Prediction

In recent years, numerous studies have highlighted the pivotal importance of miRNAs in personalized healthcare, showcasing broad application prospects. miRNAs hold significant potential in disease diagnosis, prognosis assessment, and therapeutic target discovery, making them an integral part of precision medicine. They are expected to enable precise disease subtyping and risk prediction, thereby advancing the development of precision medicine. GNNs, a class of deep learning architectures tailored for graph data analysis, have greatly facilitated the advancement of miRNA-disease association prediction algorithms. However, current methods often fall short in leveraging network node information, particularly in utilizing global information while neglecting the importance of local information. Effectively harnessing both local and global information remains a pressing challenge. To tackle this challenge, we propose an innovative model named DGNMDA. Initially, we constructed various miRNA and disease similarity networks based on authoritative databases. Subsequently, we creatively design a dual heterogeneous graph neural network encoder capable of efficiently learning feature information between adjacent nodes and similarity information across the entire graph. Additionally, we develop a specialized fine-grained multi-layer feature interaction gating mechanism to integrate outputs from the neural network encoders to identify novel associations connecting miRNAs with diseases. We evaluate our model using 5-fold cross-validation and real-world disease case studies, based on the HMDD V3.2 dataset. Our method demonstrates superior performance compared to existing approaches in various tasks, confirming the effectiveness and potential of DGNMDA as a robust method for predicting miRNA-disease associations.

Dynamic graph spatial-temporal dependence information extraction for remaining useful life prediction of rolling bearings

As a powerful tool for learning high-dimensional data representation, graph neural networks (GNN) have been applied to predict the remaining useful life (RUL) of rolling bearings. Existing GNN-based RUL prediction methods predominantly rely on constant pre-constructed graphs. However, the degradation of bearings is a dynamic process, and the dependence information between features may change at different moments of degradation. This article introduces a method for RUL prediction based on dynamic graph spatial-temporal dependence information extraction. The raw signal is segmented into multiple periods, and multiple features of each period data are extracted. Then, the correlation coefficient analysis is conducted, and the feature connection graph of each period is constructed based on different analytical results, thereby dynamically mapping the degradation process. The graph data is fed into graph convolutional networks (GCN) to extract spatial dependence between the graph node features in different periods. To make up for the shortcomings of GCN in temporal dependence extraction, the TimesNet module is introduced. TimesNet considers the two-dimensional changes of time series data and can extract the temporal dependence of graph data within and between different time cycles. Experimental results based on the PHM2012 dataset show that the average RUL prediction error of the proposed method is 17.4%, outperforming other comparative methods.

Exploring concentration-dependent transport properties on an unsteady Riga plate by incorporating thermal radiation with activation energy and gyrotactic microorganisms

Abstract The aim of this study is to examine the entropy generation (EG) associated with the transfer of mass and heat in a concentration-dependent fluid with thermal radiation and activation energy, specifically in the context of an unsteady Riga Plate with gyrotactic microorganism. It is important to solve the ordinary differential equations generated from the controlling partial differential equations using Lie symmetry scaling to verify their quality and reliability. The system’s anticipated physical behavior is compared to Mathematica’s Runge–Kutta–Fehlberg numerical solution. Source parameters are essential for validation since they offer accurate results. Methodically change these values as a percentage to determine how they affect the unsteady fluid’s density, mass, and heat transfer over the Riga plate. Velocity, temperature, nanoparticle concentration and microorganism concentration profiles decrease with varying values of the unsteadiness parameter. EG increases with increasing values of concentration difference, thermal radiation, and Reynold number parameters. The Nusselt number experiences a 26.11% rise as a result of radiation when the unsteadiness parameter is A = − 0.25 A=-0.25 , in comparison with the scenario without radiation. Mass transfer upsurges with increasing values of the Brownian motion parameter and reduces with increasing values of thermophoresis parameter. To verify our conclusions, we compare calculated data, specifically the skin friction factor, to theoretical predictions. Tabular and graphical data can show how physical limits affect flow characteristics.

Efficient generation of grids and traversal graphs in compositional spaces towards exploration and path planning

AbstractDiverse disciplines across science and engineering deal with problems related to compositions, which exist in non-Euclidean simplex spaces, rendering many standard tools inaccurate or inefficient. This work explores such spaces conceptually in the context of materials discovery, quantifies their computational feasibility, and implements several essential methods specific to simplex spaces through a new high-performance open-source library . Most significantly, we derive and implement an algorithm for constructing a novel n-dimensional simplex graph data structure, containing all discretized compositions and possible neighbor-to-neighbor transitions. Critically, no distance or neighborhood calculations are performed, instead leveraging pure combinatorics and order in procedurally generated simplex grids, keeping the algorithm $${\mathcal{O}}(N)$$ O ( N ) , with minimal memory, enabling rapid construction of graphs with billions of transitions in seconds. Additionally, we demonstrate how such graph representations can be combined to homogeneously express complex path-planning problems, while facilitating efficient deployment of existing high-performance gradient descent, graph traversal, and other optimization algorithms.

Characterizing and Understanding HGNN Training on GPUs

Owing to their remarkable representation capabilities for heterogeneous graph data, Heterogeneous Graph Neural Networks (HGNNs) have been widely adopted in many critical real-world domains such as recommendation systems and medical analysis. Prior to their practical application, identifying the optimal HGNN model parameters tailored to specific tasks through extensive training is a time-consuming and costly process. To enhance the efficiency of HGNN training, it is essential to characterize and analyze the execution semantics and patterns within the training process to identify performance bottlenecks. In this study, we conduct a comprehensive quantification and in-depth analysis of two mainstream HGNN training scenarios, including single-GPU and multi-GPU distributed training. Based on the characterization results, we reveal the performance bottlenecks and their underlying causes in different HGNN training scenarios and propose optimization guidelines from both software and hardware perspectives.

Privacy-preserving local clustering coefficient query on structured encrypted graphs

Graphs and graph databases serve as the fundamental building blocks for various network structures. In real-world network scenarios, nodes often aggregate due to their approximate organizational associations with each other. The local clustering coefficient, which evaluates the proximity of nodes within a graph, plays an important role in quantifying the structural properties of graphs in scrutinizing network robustness and understanding its intricate dynamics. Despite the growing popularity of easily accessible cloud services among small and medium-sized enterprises as well as individuals, the potential risk of data privacy disclosure when outsourcing large graphs to third-party servers is increasing. It is vital to explore a technique for executing queries on encrypted graph data. In this paper, we propose a structured encryption scheme to achieve privacy-preserving local clustering coefficient query (STE-CC) on the outsourced encrypted graphs. To calculate the clustering coefficient, we design the PSIsum protocol to sum the number of intersections, in which the basic private set intersection (PSI) protocol combines Bloom filter (BF) and garbled Bloom filter (GBF) to perform the private matching for counting the number of common neighbors. When configured with appropriate parameters, it can achieve no false negatives and negligible false positives. Finally, the security analysis and experimental evaluation on real-world graph data substantiate the effectiveness and efficiency of our approach.

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.

Knowledge Graph for Solubility Big Data: Construction and Applications

Knowledge Graph for Solubility Big Data: Construction and Applications

Eye-movement-prompted large image captioning model

Pretrained large vision-language models have shown outstanding performance on the task of image captioning. However, owing to the insufficient decoding of image features, existing large models sometimes lose important information, such as objects, scenes, and their relationships. In addition, the complex ”black-box” nature of these models makes their mechanisms difficult to explain. Research shows that humans learn richer representations than machines do, which inspires us to improve the accuracy and interpretability of large image captioning models by combining human observation patterns. We built a new dataset, called saliency in image captioning (SIC), to explore relationships between human vision and language representation. One thousand images with rich context information were selected as image data of SIC. Each image was annotated with five caption labels and five eye-movement labels. Through analysis of the eye-movement data, we found that humans efficiently captured comprehensive information for image captioning during their observations. Therefore, we propose an eye-movement-prompted large image captioning model, which is embedded with two carefully designed modules: the eye-movement simulation module (EMS) and the eye-movement analyzing module (EMA). EMS combines the human observation pattern to simulate eye-movement features, including the positions and scan paths of eye fixations. EMA is a graph neural network (GNN) based module, which decodes graphical eye-movement data and abstracts image features as a directed graph. More accurate descriptions can be predicted by decoding the generated graph. Extensive experiments were conducted on the MS-COCO and NoCaps datasets to validate our model. The experimental results showed that our network was interpretable, and could achieve superior results compared with state-of-the-art methods, i.e., 84.2% BLEU-4 and 145.1% CIDEr-D on MS-COCO Karpathy test split, indicating its strong potential for use in image captioning.

Variational spatial-temporal graph attention network for state monitoring and forecasting

With the popularity of smart devices, there has been extensive collection of data structured in graphs. Spatial graphs, in particular, have garnered significant interest owing to their diverse range of applications. In this paper, we focus on applying spatial graph model for state monitoring and forecasting, with special attention to transportation systems. In current spatial modeling approaches, understanding the structure of a spatial graph usually relies on the analysis of gathered spatial–temporal data. However, spatial graph data in real world often contains temporary factors that are not easily detected. In this paper, we propose a novel variational inference-based model VISTG, which integrates such dynamics into spatial–temporal learning of spatial graph modeling. Specifically, VISTG is primarily composed of several spatial–temporal learning blocks, each encompassing both temporal and spatial learning layers. The temporal learning layer is crafted to characterize the distributions of latent factors using a variational inference-based model, aiming to capture the dynamics within the data. Subsequently, the spatial learning layer utilizes graph attention networks to describe the correlation among nodes. Additionally, an adaptive fusion module is implemented to equalize the impact of diverse temporal patterns. Finally, comprehensive experiments are carried out on two real-world datasets. The results affirm the efficacy of our model.

GPS Tracking and Indoor Navigation System

This project is an idea of developing an application based on a navigation system formed by Graph Data Structure. An indoor positioning system (IPS) can be thought of as a global positioning system (GPS) for indoor venues. While GPS is used outdoors, IPS can detect real-time locations to accurately determine the coordinates of people or assets inside a building. These coordinates are typically represented visually by a blue dot on a digital indoor map to provide additional context to users and features such as wayfinding. The project aims to have certain objectives such as a tracking system, navigation system that is inside a small area of a building. This idea aims to track the location of any particular device within a building and show a navigation to that device in an efficient manner. Not only tracking of devices but also solving a problem of finding a particular room in the building through data of different paths that is stored. This project comes under the intelligent system design and development domain and falls under the sector of analysing and designing hardware, software and operating systems for computers. The main objective of this system design is to navigate through the area and find places to reach with ease and as fast as possible, it is helpful for people who are new at that place and even finding a user in that particular vicinity by live tracking technique.

A graph convolutional neural network model based on fused multi-subgraph as input and fused feature information as output

The graph convolution neural network (GCN)-based node classification model tackles the challenge of classifying nodes in graph data through learned feature representations. However, most existing graph neural networks primarily focus on the same type of edges, which might not accurately reflect the intricate real-world graph structure. This paper introduces a novel graph neural network model, MF-GCN, which integrates subgraphs with various edge types as input and combines feature information from each graph convolutional neural network layer to produce the final output. This model learns node feature representations by separately feeding subgraphs with different edge types into the graph convolutional layer. It then computes the weight vectors for fusing various edge type subgraphs based on the learned node features. Additionally, to efficiently extract feature information, the outputs of each graph convolution layer, without an activation function, are weighted and summed to obtain the final node features. This approach resolves the challenges of determining fusion weights and effectively extracting feature information during subgraph fusion. Experimental results show that the proposed model significantly improves performance on all three datasets, highlighting its effectiveness in node representation learning tasks.

Public Policies for the Energy Efficiency of Buildings in Mexico

In Latin America, the energy crisis has worsened due to the dependence on energy services and fossil fuel imports from highly industrialized countries at prices established by the international market; this is particularly relevant to the construction industry, which presents a significant deficit in optimal energy consumption. Hence, some governments have established public policies to maximize the efficiency of these services and, at the same time, minimize the carbon footprint. In this research study, we reviewed the public policies, strategies, and incentives for energy efficiency (EE) implementation in the residential sector established by the Mexican government. A scoping review methodology was chosen and implemented in the following steps: 1. Research inquiry identification. 2. Determination of the relevant literature and studies. 3. The literature selection. 4. Data graphing. 5. Results collection, overview, and submission. In this systematic review, we identified five mandatory standards (NOM-008-ENER-2001, NOM-009-ENER-2014, NOM-018-ENER-2011, NOM-020-ENER-2011, and NOM-024-ENER-2012), six optional standards, four strategies (Green Mortgage, Integral Sustainable Improvement in Existing Housing, ECOCASA, and NAMA), and three kinds of incentives (green bonds, credit and interest rates (Green Mortgage, FIDE, and Ecocasa), and taxes (Income Tax Reduction)). As a result of the implementation of the above, as of December 2020, NAMA financed 5106 developers of 38 projects in 15 states; contributed to a reduction of 126,779 tons of CO2; and aided 19,913 people. From 2013 to December 2023, EcoCasa subsidized 71,440 households for a total of 224 projects in 25 states; contributed to a reduction of 2.6 million tons of CO2; aided 285,760 Mexicans; and issued EcoCasa certificates for 3,473,556 m2. The results of the EE indicators in residential buildings showed an increase in the housing unit number as well as an increase in household appliances, with those based on power consumption prevailing. The residential sector ranks third in power consumption in Mexico, consuming an estimated 790 pj, of which 76% corresponds to thermal energy and 24% to electric power. Among countries in Latin America and the Caribbean, Mexico has achieved an Energy Transition Index of 62%.

An automated data collection process for constructing graph data relying on LLMs

This paper introduces a process that is designed to harvest data automatically from a variety of online sources. The core of this process lies in its data-handling techniques, which include drawing, cleaning, deduplicating, extracting, and categorizing of raw data to convert unstructured data into a structured format represented and imported in a graph database. The data extraction step utilizes Large Language Model (LLMs) for Named Entity Recognition (NER). A case study on deploying course data collection illustrates the enhancements brought about by this automation, showcasing improvements in the accuracy, completeness, and timeliness of updates in the course data. An evaluation carried out on the extraction and matching methods shows that the F1-score and precision rates are high. Overall, this study contributes to advancement of the field by providing a methodology for automating the collection and processing of online data sources, significantly improving the quality of data collection from online sources.