Representative Network Models Research Articles

Dynamic traffic network representation model for improving the prediction performance of passenger flow for mass rapid transit

Accurate machine learning predictions of passenger flow data for mass rapid transit (MRT) systems can considerably improve operational efficiency by enabling better allocation of train and human resources. However, such predictions are challenging because MRT networks have complex structures with route dependence and transfer stations. Although the static state of an MRT network has been computed in previous studies, a comprehensive understanding of an MRT network requires characterizing its dynamics. Therefore, this paper proposes a dynamic traffic network representation (DTNR) model that captures station features from historical traffic flows and geographical information of MRT stations. Furthermore, a multilevel attention network (MLAN) model is proposed to predict MRT passenger flow as a downstream task following the pretraining of the DTNR model. The experimental results of this study indicate that the developed DTNR and MLAN models can accurately predict MRT passenger flow. These models are widely applicable to different MRT systems and passenger flow situations, making them a valuable tool for transportation planners and operators.

Research on the dynamic spread of information in social networks based on relationship strength theory and feedback mechanism

Introduction: Studying the main factors and related paths of rumor propagation contributes to the precise governance of rumor information in social networks. Most existing network representation learning methods do not fit with real-world information propagation networks well, and the network cannot effectively model the temporal characteristics and dynamic evolution features of rumor information propagation.Methods: Our study proposes a new dynamic network representation model for information propagation. Additionally, the study introduces a feedback mechanism where the latest node representations are fed back to neighboring nodes.Results: The method solves the problem of delayed network representation and improves network representation performance.Discussion: We conducted experimental simulations, and the results indicate that a higher level of trust contributes to stable group relationships and converging opinions, reducing the likelihood of opinion dispersion. Furthermore, novelty of topics, and interactivity between users, and opinion leaders exhibit distinct characteristics in guiding public opinion. The viewpoint evolution of the newly constructed dynamic network representation model aligns more closely with viewpoint evolution in real-world social networks.

A multi-source molecular network representation model for protein–protein interactions prediction

The prediction of potential protein–protein interactions (PPIs) is a critical step in decoding diseases and understanding cellular mechanisms. Traditional biological experiments have identified plenty of potential PPIs in recent years, but this problem is still far from being solved. Hence, there is urgent to develop computational models with good performance and high efficiency to predict potential PPIs. In this study, we propose a multi-source molecular network representation learning model (called MultiPPIs) to predict potential protein–protein interactions. Specifically, we first extract the protein sequence features according to the physicochemical properties of amino acids by utilizing the auto covariance method. Second, a multi-source association network is constructed by integrating the known associations among miRNAs, proteins, lncRNAs, drugs, and diseases. The graph representation learning method, DeepWalk, is adopted to extract the multisource association information of proteins with other biomolecules. In this way, the known protein–protein interaction pairs can be represented as a concatenation of the protein sequence and the multi-source association features of proteins. Finally, the Random Forest classifier and corresponding optimal parameters are used for training and prediction. In the results, MultiPPIs obtains an average 86.03% prediction accuracy with 82.69% sensitivity at the AUC of 93.03% under five-fold cross-validation. The experimental results indicate that MultiPPIs has a good prediction performance and provides valuable insights into the field of potential protein–protein interactions prediction. MultiPPIs is free available at https://github.com/jiboyalab/multiPPIs.

DBSCAN-based energy users clustering for performance enhancement of deep learning model

Background: Due to rapid progress in the fields of artificial intelligence, machine learning and deep learning, the power grids are transforming into Smart Grids (SG) which are versatile, reliable, intelligent and stable. The power consumption of the energy users is varying throughout the day as well as in different days of the week. Power consumption forecasting is of vital importance for the sustainable management and operation of SG. Methodology: In this work, the aim is to apply clustering for dividing a smart residential community into several group of similar profile energy user, which will be effective for developing and training representative deep neural network (DNN) models for power load forecasting of users in respective groups. The DNN models is composed of convolutional neural network (CNN) followed by LSTM layers for feature extraction and sequence learning respectively. The DNN For experimentation, the Smart Grid Smart City (SGSC) project database is used and its energy users are grouped into various clusters. Results: The residential community is divided into four groups of customers based on the chosen criterion where Group 1, 2, 3 and 4 contains 14 percent, 22 percent, 19 percent and 45 percent users respectively. Almost half of the population (45 percent) of the considered residential community exhibits less than 23 outliers in their electricity consumption patterns. The rest of the population is divided into three groups, where specialized deep learning models developed and trained for respective groups are able to achieve higher forecasting accuracy. The results of our proposed approach will assist researchers and utility companies by requiring fewer specialized deep-learning models for accurate forecasting of users who belong to various groups of similar-profile energy consumption.

Multi-view learning-based heterogeneous network representation learning

Network representation learning is an important tool for extracting latent features from heterogeneous networks to enhance downstream analysis tasks. However, for heterogeneous networks in the era of big data, their heterogeneity, unseen network noises, latent structural and semantic features, and the coupling degree between the two features are increasing. These unignorable changes lead to the difficulties of extracting latent features and the demand for further optimization of existing embedding models. In this paper, an unsupervised multi-view learning based heterogeneous network representation model is proposed, called MVHNR. Firstly, to simplify the difficulty of feature learning, a novel view generation strategy is designed to divide the heterogeneous network into multiple sub-views that contain only one semantic feature and its related structural features. Then, to improve the accuracy and robustness of feature learning with network noises, a novel adjacency matrix strategy is constructed for each view, and a novel adversarial learning strategy is used to both learn the local semantic and structural features in each view. Finally, to aggregate the local features of all views, a multi-headed attention network is used to generate a global feature vector of each node. Extensive experimental results on three tasks and five public networks show the advantages of our MVHNR model.

Predicting the effects of mutations on protein solubility using graph convolution network and protein language model representation.

Solubility is one of the most important properties of protein. Protein solubility can be greatly changed by single amino acid mutations and the reduced protein solubility could lead to diseases. Since experimental methods to determine solubility are time-consuming and expensive, in-silico methods have been developed to predict the protein solubility changes caused by mutations mostly through protein evolution information. However, these methods are slow since it takes long time to obtain evolution information through multiple sequence alignment. In addition, these methods are of low performance because they do not fully utilize protein 3D structures due to a lack of experimental structures for most proteins. Here, we proposed a sequence-based method DeepMutSol to predict solubility change from residual mutations based on the Graph Convolutional Neural Network (GCN), where the protein graph was initiated according to predicted protein structure from Alphafold2, and the nodes (residues) were represented by protein language embeddings. To circumvent the small data of solubility changes, we further pretrained the model over absolute protein solubility. DeepMutSol was shown to outperform state-of-the-art methods in benchmark tests. In addition, we applied the method to clinically relevant genes from the ClinVar database and the predicted solubility changes were shown able to separate pathogenic mutations. All of the data sets and the source code are available at https://github.com/biomed-AI/DeepMutSol.

Novel network representation model for improving controllability processes on temporal networks

Abstract Temporal networks are known as the most important tools for representing and storing dynamic systems. This type of network accurately demonstrates all the dynamic changes that occur in a dynamic system. In different applications of dynamic systems, different representation of network models has been used to represent temporal networks. In the last decade, controllability in dynamic systems has become one of the most important challenges in this field. Controllability means the transfer of the network from an initial state to a desired final state in a certain period of time. The most common representation of network model used in control processes is the layered model. But this model has a high overhead, and on the other hand, it slows down the network control processes. In this article, we have proposed a new model for storing and representing temporal networks, which uses a tree structure to save all dynamics of network. Considering that in the proposed model only essential network control information is stored, this model has a very low data overhead compared to the layered model, and this makes the control processes run at a higher speed.

Retrieving Atmospheric Gas Profiles Using FY-3E/HIRAS-II Infrared Hyperspectral Data by Neural Network Approach

The observed radiation data from the second-generation Hyperspectral Infrared Atmospheric Sounder (HIRAS-II) on the Fengyun-3E (FY-3E) satellite contain useful vertical atmosphere information which can distinguish and retrieve vertical profiles of atmospheric gas components including ozone (O3), carbon monoxide (CO), and methane (CH4). This paper utilizes FY-3E/HIRAS-II observational data to optimize each gas channel using the improved Optimal Sensitivity Profile method (OSP) channel algorithm and establishes a typical convolutional neural network model (CNN) and a representative U-shaped network model (UNET) with deep features and shallow feature links to perform atmospheric profile retrieval calculations of O3, CO, and CH4. We chose the clear sky data of the Indian and its southern seas in December 2021 and January 2022, with reanalysis data from European Center for Medium-Range Weather Forecasts Reanalysis v5 (ERA5) and European Center for Medium-Range Weather Forecasts Atmospheric Composition Reanalysis v4 (EAC4) serving as the reference values. The retrieval outcomes were then compared against advanced numerical forecast models including the Whole Atmosphere Community Climate Model (WACCM), Global Forecast System (GFS), and satellite products from an Atmospheric Infrared Sounder (AIRS) and Infrared Atmospheric Sounding Interferometer (IASI). Experimental results show that the generalization ability and retrieval accuracy of CNN are slightly higher compared with UNET. For O3 profile retrieval, the mean percentage error (MPE) of the whole layers for CNN and UNET data in relation to ERA5 data was less than 8%, while the root-mean-square error (RMSE) was below 1.5 × 10−7 kg/kg; for CH4 profile retrieval, the MPE of the whole layers for CNN and UNET data in relation to EAC4 data was less than 0.7%, while the RMSE was below 1.5 × 10−8 kg/kg. The retrieval of O3 and CH4 are resulted in a significant improvement compared to the forecast data and satellite products in most pressure levels; for CO profile retrieval, the MPE of the whole layers for CNN and UNET data in relation to EAC4 data was less than 11%, while the RMSE was below 4 × 10−8 kg/kg. The error of the CO retrieval results was higher than that of the forecast data at the pressure level of 200~500 hPa and lower than that of similar satellite products with most pressure levels. The experiments indicated that the neural network method effectively determines the atmospheric gas profiles using infrared hyperspectral data, exhibiting a positive performance in accuracy and retrieval speed.

ENSURING DATA UNIQUENESS IN SEMANTIC NETWORKS

The article gives a brief description of knowledge representation models. Atoms of meaning (basic, minimal informational units) combined with each other to express a common meaning represent knowledge (data about data, metadata). It is shown that most of the existing knowledge representation models are based on the network representation model. A method is proposed to ensure the uniqueness (originality) of a set of data underlying the network model of knowledge representation. To ensure the uniqueness of knowledge representation by a set of data, the article proposes to use the main theorem of arithmetic: data are denoted by simple numbers (identifiers); when multiplying among themselves several prime numbers (data identifiers) in their totality conveying the aggregate semantic meaning, a natural number is obtained, which is an identifier of knowledge. The use of the basic theorem of arithmetic also provides a formalization of important data properties: internal interpretability, structuredness, connectivity, semantic metrics and activity. The presence of these properties in the data indicates that it is already data over data, or in other words, knowledge. In certain cases, this can reduce the computational complexity of the algorithm of the linguistic processor.

Prediction of two-phase flow properties for digital sandstones using 3D convolutional neural networks

Multiphase fluid flow within porous media is of great importance in a wide range of environmental and industrial fields, such as CO2 sequestration, geothermal systems, fuel cells, enhanced oil recovery, and groundwater remediation. Pore-scale modeling is a promising tool to estimate macroscopic multiphase flow properties from micro-scale images of porous materials; however, it is a complex, time-consuming procedure and would be highly resource-intensive in the case of direct numerical simulation. We present a framework that consists of (1) extraction of sub-samples from rock images, (2) computation of relative permeability and capillary pressure from two-phase pore network modeling with respect to their contact angle and interfacial tensions, (3) training a convolutional neural network (CNN), and (4) validation with unseen datasets and parameters. 500 sub-samples of grayscale and binary types are extracted from images of 12 different sandstones. Relative permeability, capillary pressure, and residual saturation of binary sub-samples are computed from two-phase fluid flow simulation in their representative pore network models The proposed CNN model is trained with grayscale images, eliminating the need for image processing. The results demonstrate a good agreement between CNN predictions and simulation results, and the computational time was reduced by several orders of magnitude.

Dynamic Student Data Management Using Resource Optimization Technology in Higher Education Platforms

This study combines intelligent resource optimization technology to build a dynamic student data management model and suggests a fuzzy hierarchical network representation model based on isomorphism and homogeneity in order to increase the effectiveness of dynamic student data management in colleges and universities. The important semantics of nodes in the network are also captured in this study using fuzzy k-kernel decomposition as a technique for multigranularity partitioning. Based on the SIR model, FHNE compares the production of the sequence to the process of information transmission in the random walk stage, which increases the node sequence’s accuracy. According to the research, the dynamic student data management system that is used in the higher education platform that is suggested in this study can significantly increase the effectiveness of managing student data.

Deep Learning Models for Single-Channel Speech Enhancement on Drones

Speech enhancement for drone audition is made challenging by the strong ego-noise from the rotating motors and propellers, which leads to extremely low signal-to-noise ratios (e.g. SNR < -15 dB) at onboard microphones. In this paper, we extensively assess the ability of single-channel deep learning approaches to ego-noise reduction on drones. We train twelve representative deep neural network (DNN) models, covering three operation domains (time-frequency magnitude domain, time-frequency complex domain and end-to-end time domain) and three distinct architectures (sequential, encoder-decoder and generative). We critically discuss and compare the performance of these models in extremely low-SNR scenarios, ranging from -30 to 0 dB. We show that time-frequency complex domain and UNet encoder-decoder architectures outperform other approaches on speech enhancement measures while providing a good trade-off with other criteria, such as model size, computation complexity and context length. Specifically, the best-performing model is DCUNet, a UNet model operating in the time-frequency complex domain, which, at input SNR -15 dB, improves ESTOI from 0.1 to 0.4, PESQ from 1.0 to 1.9 and SI-SDR from -15 dB to 3.7 dB. Based on the insights drawn from these findings, we discuss future research in drone ego-noise reduction.

Network-level signal predictive control with real-time routing information

This paper presents a framework for signalized road network predictive optimization using real-time routing information from connected vehicles (CVs). An important feature of the real-time routing information is the ability of CVs to broadcast the target routes they expect to travel through to the infrastructure in real time while assuming that a majority of the CVs can provide their target routes. A fully movement-level network representation model is proposed to easily describe the traffic state and demand of the signalized network. The problem is formulated as a mixed integer linear programming model to predict the movement-level network dynamics, which is solved in real time to optimize phase-free movement signal timings. The objective is to maximize the network throughput within the prediction horizon while avoiding queue spillbacks. A decentralized solution algorithm is developed to decompose the network-level problem into intersection-level subproblems, thereby reducing computational complexity. Simulation experiments validate the advantages of the proposed framework over TRANSYT schemes and max pressure-based control strategy in various scenarios. Sensitivity analysis shows the control performance under different traffic demand levels and penetration rates of the target route information. Comparisons in prediction accuracy, control performance, and computational efficiency between centralized and decentralized solutions of the proposed model are also conducted. This study explores the application of real-time routing information as a potential type of data for the network-level predictive signal optimization in the future CV environment.

Enhancement of automatic classification of arcus senilis-nonarcus senilis using convolutional neural networ

Cholesterol is a type of lipid found in the human body and is susceptible to abnormalities. It can be detected via lipid profiling through blood sampling. In addition, cholesterol can be detected through the presence of a "sodium ring" in the eye iris called the corneal arcus (CA), presenting a new preliminary detection method that is less invasive. Therefore, this paper proposed a non-invasive method in detecting cholesterol based on convolutional neural network (CNN) model representation using 300 normal and 300 abnormal iris images from UBIRIS and medical web images. In this work, contrast-limited adaptive histogram (CLAHE) and unsharp masking process was applied first on CA images to enhance the quality of CA images. To detect the CA images, the dataset was trained and tested using three pre-trained CNN architectures; one is created from scratch, another are Resnet-50 and VGG-19 architectures that were fine-tuned to the CA images. The best result was exhibited by proposed pre-trained CNN model created from scratch with 10-fold cross-validation that produced high average detection accuracy at 98.81%. Thus, deeper network implementation is recommended in the future to further improve CA localization for optometrists used in their daily clinical tasks in detecting cholesterol.

Connectivity concepts in neuronal network modeling.

Sustainable research on computational models of neuronal networks requires published models to be understandable, reproducible, and extendable. Missing details or ambiguities about mathematical concepts and assumptions, algorithmic implementations, or parameterizations hinder progress. Such flaws are unfortunately frequent and one reason is a lack of readily applicable standards and tools for model description. Our work aims to advance complete and concise descriptions of network connectivity but also to guide the implementation of connection routines in simulation software and neuromorphic hardware systems. We first review models made available by the computational neuroscience community in the repositories ModelDB and Open Source Brain, and investigate the corresponding connectivity structures and their descriptions in both manuscript and code. The review comprises the connectivity of networks with diverse levels of neuroanatomical detail and exposes how connectivity is abstracted in existing description languages and simulator interfaces. We find that a substantial proportion of the published descriptions of connectivity is ambiguous. Based on this review, we derive a set of connectivity concepts for deterministically and probabilistically connected networks and also address networks embedded in metric space. Beside these mathematical and textual guidelines, we propose a unified graphical notation for network diagrams to facilitate an intuitive understanding of network properties. Examples of representative network models demonstrate the practical use of the ideas. We hope that the proposed standardizations will contribute to unambiguous descriptions and reproducible implementations of neuronal network connectivity in computational neuroscience.

Finding Asymptomatic Spreaders in a COVID-19 Transmission Network by Graph Attention Networks.

In the COVID-19 epidemic the mildly symptomatic and asymptomatic infections generate a substantial portion of virus spread; these undetected individuals make it difficult to assess the effectiveness of preventive measures as most epidemic prevention strategies are based on the detected data. Effectively identifying the undetected infections in local transmission will be of great help in COVID-19 control. In this work, we propose an RNA virus transmission network representation model based on graph attention networks (RVTR); this model is constructed using the principle of natural language processing to learn the information of gene sequence and using a graph attention network to catch the topological character of COVID-19 transmission networks. Since SARS-CoV-2 will mutate when it spreads, our approach makes use of graph context loss function, which can reflect that the genetic sequence of infections with close spreading relation will be more similar than those with a long distance, to train our model. Our approach shows its ability to find asymptomatic spreaders both on simulated and real COVID-19 datasets and performs better when compared with other network representation and feature extraction methods.

Discovering miRNAs Associated With Multiple Sclerosis Based on Network Representation Learning and Deep Learning Methods.

Identifying biomarkers of Multiple Sclerosis is important for the diagnosis and treatment of Multiple Sclerosis. The existing study has shown that miRNA is one of the most important biomarkers for diseases. However, few existing methods are designed for predicting Multiple Sclerosis-related miRNAs. To fill this gap, we proposed a novel computation framework for predicting Multiple Sclerosis-associated miRNAs. The proposed framework uses a network representation model to learn the feature representation of miRNA and uses a deep learning-based model to predict the miRNAs associated with Multiple Sclerosis. The evaluation result shows that the proposed model can predict the miRNAs associated with Multiple Sclerosis precisely. In addition, the proposed model can outperform several existing methods in a large margin.

Flow capacity characterization of unconventional natural gas bearing rocks using digital rock physics: A comparison between advection and diffusion

Experimental flow capacity characterization of unconventional rocks is a nontrivial task, and the relative contribution of advective vs. diffusive flow mechanisms remains illusive. Herein we apply a digital rock physics workflow to systematically compare the advection and diffusion behavior of supercritical methane in two 3D reconstructed unconventional shale and coal core samples. Nitrogen sorption experiments and scanning electron microscope observations are combined to inform the construction of representative pore network models. For advection modeling, we consider the nanometer-scale confinement effect and pore shape effect, while for diffusion modeling the contributions of Fick diffusion, transitional diffusion, and Knudsen diffusion are considered separately based on the Knudsen number. The effect of pore size, shape, pressure, and temperature on methane flow capacity in a single nanopore is first studied. Then the apparent permeability and effective diffusivity of the two reconstructed nanoporous media are calculated and compared. We demonstrate that at reservoir pressure and temperature conditions, diffusion contributes to mass flux (above 10%) only in micropores and small mesopores (< 10 nm). For most realistic unconventional rocks, the contribution of bulk diffusion can be neglected when estimating the flow capacity.

Computation of the Protein Conformational Transition Pathway on Ligand Binding by Linear Response-Driven Molecular Dynamics.

While extremely important for relating the protein structure to its biological function, determination of the protein conformational transition pathway upon ligand binding is made difficult due to the transient nature of intermediates, a large and rugged conformational space, and coupling between protein dynamics and ligand-protein interactions. Existing methods that rely on prior knowledge of the bound (holo) state structure are restrictive. A second concern relates to the correspondence of intermediates obtained to the metastable states on the apo → holo transition pathway. Here, we have taken the protein apo structure and ligand-binding site as only inputs and combined an elastic network model (ENM) representation of the protein Hamiltonian with linear response theory (LRT) for protein-ligand interactions to identify the set of slow normal modes of protein vibrations that have a high overlap with the direction of the protein conformational change. The structural displacement along the chosen direction was performed using excited normal modes molecular dynamics (MDeNM) simulations rather than by the direct use of LRT. Herein, the MDeNM excitation velocity was optimized on-the-fly on the basis of its coupling to protein dynamics and ligand-protein interactions. Thus, a determined set of structures was validated against crystallographic and simulation data on four protein-ligand systems, namely, adenylate kinase-di(adenosine-5')pentaphosphate, ribose binding protein-β-d-ribopyranose, DNA β-glucosyltransferase-uridine-5'-diphosphate, and G-protein α subunit-guanosine-5'-triphosphate, which present important differences in protein conformational heterogeneity, ligand binding mechanism, viz. induced-fit or conformational selection, extent, and nonlinearity in protein conformational changes upon ligand binding, and presence of allosteric effects. The obtained set of intermediates was used as an input to path metadynamics simulations to obtain the free energy profile for the apo → holo transition.

Network representation learning method embedding linear and nonlinear network structures

With the rapid development of neural networks, much attention has been focused on network embedding for complex network data, which aims to learn low-dimensional embedding of nodes in the network and how to effectively apply learned network representations to various graph-based analytical tasks. Two typical models exist namely the shallow random walk network representation method and deep learning models such as graph convolution networks (GCNs). The former one can be used to capture the linear structure of the network using depth-first search (DFS) and width-first search (BFS), whereas Hierarchical GCN (HGCN) is an unsupervised graph embedding that can be used to describe the global nonlinear structure of the network via aggregating node information. However, the two existing kinds of models cannot simultaneously capture the nonlinear and linear structure information of nodes. Thus, the nodal characteristics of nonlinear and linear structures are explored in this paper, and an unsupervised representation method based on HGCN that joins learning of shallow and deep models is proposed. Experiments on node classification and dimension reduction visualization are carried out on citation, language, and traffic networks. The results show that, compared with the existing shallow network representation model and deep network model, the proposed model achieves better performances in terms of micro-F1, macro-F1 and accuracy scores.