Multilayer Network Model Research Articles

Characterizing the role of human behavior in the effectiveness of contact-tracing applications.

Although numerous countries relied on contact-tracing (CT) applications as an epidemic control measure against the COVID-19 pandemic, the debate around their effectiveness is still open. Most studies indicate that very high levels of adoption are required to stop disease progression, placing the main interest of policymakers in promoting app adherence. However, other factors of human behavior, like delays in adherence or heterogeneous compliance, are often disregarded. To characterize the impact of human behavior on the effectiveness of CT apps we propose a multilayer network model reflecting the co-evolution of an epidemic outbreak and the app adoption dynamics over a synthetic population generated from survey data. The model was initialized to produce epidemic outbreaks resembling the first wave of the COVID-19 pandemic and was used to explore the impact of different changes in behavioral features in peak incidence and maximal prevalence. The results corroborate the relevance of the number of users for the effectiveness of CT apps but also highlight the need for early adoption and, at least, moderate levels of compliance, which are factors often not considered by most policymakers. The insight obtained was used to identify a bottleneck in the implementation of several apps, such as the Spanish CT app, where we hypothesize that a simplification of the reporting system could result in increased effectiveness through a rise in the levels of compliance.

Lump wave solutions, lump-stripe soliton inelastic collision phenomena and rogue-type wave solutions for a generalized breaking soliton system in (3+1)-dimensions

Based on five types of structures of the functional multi-layer neural network model, some explicit exact solutions for a (3+1)-dimensional generalized breaking soliton system including lump waves, lump-stripe solitons and rogue-type waves are explored by using bilinear neural network method without the datasets training. The 3-dimensional plots, 2-dimensional curves and density plots of these results obtained are provided to display their dynamic characteristics. In particular, the complete processes of lump wave and stripe soliton fusion phenomena are presented. We hope the formulations established in this paper could be applied to study the exact traveling wave solutions of other integrable nonlinear evolution equations.

Robustness measurement of multiplex networks based on multiplex motifs

Multiplex network is a classic multilayer network model that can be used to model complex systems in the real world. Previous measures of network robustness mainly focused on the low-order characteristics of network topology, and the existence of motifs in the network is ignored. As simple building blocks of multiplex networks, multiplex motifs are indispensable tools to understand the structure of local networks. Therefore, from a mesoscopic perspective, we propose an index called normalized motif entropy to measure the robustness of multiplex networks under malicious attacks, and the effectiveness of our method is verified in real multiplex networks.

Investigation of herding behavior using machine learning models

PurposeThe study aims to investigate the presence of herding behavior in the stock market of UK with a special emphasis on news sentiment regarding the economy. The authors focus on the news sentiment because in the current digital era, investors take their decision making on the basis of current trends projected by news and media platforms.Design/methodology/approachFor empirical modeling, the authors use machine learning models to investigate the presence of herding behavior in UK stock market for the period starting from 2006 to 2021. The authors use support vector regression, single layer neural network and multilayer neural network models to predict the herding behavior in the stock market of the UK. The authors estimate the herding coefficients using all the models and compare the findings with the linear regression model.FindingsThe results show a strong evidence of herding behavior in the stock market of the UK during different time regimes. Furthermore, when the authors incorporate the economic uncertainty news sentiment in the model, the results show a significant improvement. The results of support vector regression, single layer perceptron and multilayer perceptron model show the evidence of herding behavior in UK stock market during global financial crises of 2007–08 and COVID’19 period. In addition, the authors compare the findings with the linear regression which provides no evidence of herding behavior in all the regimes except COVID’19. The results also provide deep insights for both individual investors and policy makers to construct efficient portfolios and avoid market crashes, respectively.Originality/valueIn the existing literature of herding behavior, news sentiment regarding economic uncertainty has not been used before. However, in the present era this parameter is quite critical in context of market anomalies hence and needs to be investigated. In addition, the literature exhibits varying results about the existence of herding behavior when different methodologies are used. In this context, the use of machine learning models is quite rare in the herding literature. The machine learning models are quite robust and provide accurate results. Therefore, this research study uses three different models, i.e. single layer perceptron model, multilayer perceptron model and support vector regression model to investigate the herding behavior in the stock market of the UK. A comparative analysis is also presented among the results of all the models. The study sheds light on the importance of economic uncertainty news sentiment to predict the herding behavior.

Applying the Multi-Layer Perceptron Neural Network Model to Predicting School Closures: An Example of Taipei City

In this study, we utilised multi-layer perceptron neural networks (MLPNNs) to assess the issue of school closures. Specifically, this study uses the MLPNN model to conduct learning assessments to identify schools that may suffer from poor management. The empirical findings are briefly summarised as follows: (1) The research shows that more than half of the private schools in this study will face bankruptcy. In Taipei City, out of the 23 existing private schools, only four operate normally, while the remaining 19 private vocational high schools require assistance. About 12 schools face severe problems in terms of poor management, accounting for 63% of the total, while seven schools had a prediction value below 50, indicating a severe problem. These schools have expressed an immediate need for government assistance. (2) According to the MLPNN model in this study, reducing the number of full-time teachers is a primary factor contributing to school closures. First, since full-time teachers are a fixed cost, the dismissal of teachers tends to be prioritised to bring down school management costs. This, in turn, reduces the teacher-student ratio. Other factors that contribute to school bankruptcy are dismissing staff and part-time teachers, reducing expenditure, and poor operational maintenance. When the above policies are implemented in schools with poor management, a vicious cycle is created, leading to the bankruptcy of schools.

A Systematic Review of Neural Network Autoregressive Model with Exogenous Input for Solar Radiation Prediction Modelling Development

Neural Network is one of the Machine Learning methods that has been applied in various Artificial Intelligence system development including solar radiation prediction modelling. Since there are multiple approaches had been developed using the Neural Network method, the study has been focusing on the development of a Multi-layer Neural Network model that can handle non-linearities and highly dynamic data. The integration of the Multi-layer Neural Network and the Non-linear Autoregressive Model with Exogenous Input (NARX) developed a compromising non-linear Neural Network model which can be applied in the modelling of solar radiation. This paper develops a systematic review of the Neural Network Autoregressive Model with Exogenous Input (NNARX) for solar radiation prediction modelling starts from the architecture and the comparative selection for the Training Function. The model is developed and analysed using MATLAB R2019a software. Results showed that the Levenberg-Marquardt Training Function performed better with the R2 value of 0.94 for training and 0.91 for testing, making it the most suitable for the NNARX in the development of solar radiation prediction modelling.

Fractional ordering of activation functions for neural networks: A case study on Texas wind turbine

Activation functions play an important role in deep learning models by introducing non-linearity to the output of a neuron, enabling the network to learn complex patterns and non-linear relationships in data and make predictions on more complex tasks. Deep learning models’ most commonly used activation functions are Purelin, Sigmoid, Tansig, Rectified Linear Unit (ReLU), and Exponential Linear Unit (ELU), which exhibit limitations such as non-differentiability, vanishing gradients, and neuron inactivity with negative values. These functions are typically defined over a finite range, and their outputs are integers or real numbers. Using fractional calculus in designing activation functions for neural networks has shown promise in improving the performance of deep learning models in specific applications. These activation functions can capture more complex non-linearities than traditional integer-order activation functions, improving performance on tasks such as image classification and time series prediction. This paper focuses on deriving and testing linear and non-linear fractional-order forms of activation functions and their variants. The linear activation function includes Purelin. In contrast, the non-linear activation functions are Binary Step, Sigmoid, Tansig, ReLU, ELU, Gaussian Error Linear Unit (GELU), Hexpo, and their variants. Besides, the standard formula has been implemented and used in developing the fractional-order linear activation function. Furthermore, various expansion series, such as Euler and Maclaurin, have been used to design non-linear fractional-order activation functions and their variants. The single- and multi-layer fractional-order neural network models have been developed using the designed fractional-order activation functions. The simulation study uses developed fractional-order neural network models for predicting the Texas wind turbine systems’ generated power. The performance of single and multi-layer fractional-order neural network models has been evaluated by changing the activation functions in the hidden layer while keeping the Purelin function constant at the output layer. Experiments on neural network models demonstrate that the designed fractional-order activation functions outperform traditional functions like Sigmoid, Tansig, ReLU, ELU, and their variants, effectively addressing limitations.

Abnormal multi-layered dynamic cortico-subcortical functional connectivity in major depressive disorder and generalized anxiety disorder

Comorbidity has been frequently observed between generalized anxiety disorder (GAD) and major depressive disorder (MDD), however, common and distinguishable alterations in the topological organization of functional brain networks remain poorly understood. We sought to determine a robust and sensitive functional connectivity marker for diagnostic classification and symptom severity prediction. Multi-layered dynamic functional connectivity including whole brain, network-node and node-node layers via graph theory and gradient analyses were applied to functional MRI resting-state data obtained from 31 unmedicated GAD and 34 unmedicated MDD patients as well as 33 age and education matched healthy controls (HC). GAD and MDD symptoms were assessed using Penn State Worry Questionnaire and Beck Depression Inventory II, respectively. Three network measures including global properties (i.e., global efficiency, characteristic path length), regional nodal property (i.e., degree) and connectivity gradients were computed. Results showed that both patient groups exhibited abnormal dynamic cortico-subcortical topological organization compared to healthy controls, with MDD > GAD > HC in degree of randomization. Furthermore, our multi-layered dynamic functional connectivity network model reached 77% diagnostic accuracy between GAD and MDD and was highly predictive of symptom severity, respectively. Gradients of functional connectivity for superior frontal cortex-subcortical regions, middle temporal gyrus-subcortical regions and amygdala-cortical regions contributed more in this model compared to other gradients. We found shared and distinct cortico-subcortical connectivity features in dynamic functional brain networks between GAD and MDD, which together can promote the understanding of common and disorder-specific topological organization dysregulations and facilitate early neuroimaging-based diagnosis.

Intrusion detection system for detecting distributed denial of service attacks using machine learning algorithms

<span>Today, the creation of more effective intrusion detection system (IDS) has become crucial due to the rise in computer malware. Ensuring the availability of the system is an important component of information security and the most important requirement of any network. Recently the machine learning algorithm (ML) has been used to improve intrusion detection over the network. It is currently necessary to release an updated version of these systems. The presented work aimed to build a reliable and accurate IDS based on ML to classify and prevent distributed denial of service attacks to protect any system working on the network from temporary or complete system failure. We presented five ML models to create the proposed distributed denial-of-services attack (DDoS)-IDS, including (decision tree, random forest, logistic regression, support vector machine, and multi-layer neural network) which were trained and evaluated using the CIC-IDS-2018 dataset. Furthermore, principal component analysis (PCA) was used to reduce the dimensionality of the dataset. According to the classification results, the proposed multi-layer neural network model reached optimal performance for detecting DDoS attacks and achieved classification accuracy at 99.9992%.</span>

Neural Network Modelling for Prediction of Coagulant Dosage

Water treatment plants (WTP) in Indonesia commonly applied conventional treatment that, among others, relies on the coagulation process to treat turbid surface water to produce clean and safe water for daily purposes. Meanwhile, standard Jar-Test methods for coagulant dosage determination are inefficient, inaccurate, and possibly will not function well for water with high variability of quality. This study aims to improve and optimize the coagulant dosing process by considering the main parameter that greatly affects the coagulation process, i.e., turbidity. For this purpose, a multilayer artificial neural network model (ANN) with an adaptive backpropagation algorithm was developed using MATLAB software. A network architecture with 1 input, 5 hidden and 1 output was implemented. For the input, 4465 turbidity data of the raw water taken from Surabaya Municipal Water Enterprise (PDAM) were used, while the output is the optimal dose obtained from the results of the Jar-test. This study resulted in an empirical model of the optimum dosage of coagulants as a function of the turbidity of the raw water. Training and testing on the given ANN resulted in R2 (determination coefficient) of 0.99444 from regression analysis with the following equation: output = 0.99 x target + 0.66. This study showed that the ANN model provides a very high success rate for coagulant dose prediction.

A product requirement development method based on multi-layer heterogeneous networks

The rapid advancement of emerging technologies in the information age has resulted in intricate, diverse, and rapidly changing user requirements, thereby reducing the effectiveness of traditional requirement development approaches. This paper proposes a multi-layer heterogeneous network-based method for product requirement development to identify key requirements in the product development process and address the complex relationships among users, requirements, products, and requirement development teams. By incorporating the theory of hesitant fuzzy into the expert decision-making process, this method constructs four single-layer networks: user attention network, expert trust network, requirement correlation network, and component combination network, which assigns realistic significance to nodes in each network layer and establishes connections between different networks, forming a multi-layer heterogeneous network model for product requirement development. Furthermore, this paper defines indicators such as layer significance and node comprehensive significance and proposes a calculation method for node comprehensive significance based on biased random walk, which enables deep integration of node’s realistic significance and topological centrality. The proposed method is applied to a data experiment of product requirement development for a specific type of new energy vehicle. By analyzing the calculation results of node comprehensive significance, high-quality recommendations are provided for the development team from the perspectives of functional requirements, product components, and user groups. Moreover, comparative experiments to assess algorithm performance demonstrate that the average accuracy of the proposed algorithm is 0.95, which is significantly higher than that of the commonly used single-layer network node centrality measurement methods and recently proposed multi-layer network node centrality measurement methods. These results reflect the advancement and applicability of the proposed method in product development scenarios.

Carbon emission characteristics of urban trip based on multi-layer network modeling

Multi-layer networks could reveal the carbon emission structure of urban traffic formed after residents choose the means and purpose of trips. In this paper, a multi-layer network model was proposed and the carbon emission characteristics of urban trips were analyzed. In addition, the carbon reduction potential assessment methods based on nodes and edge feature indexes to identify the carbon reduction areas of the trip network. An empirical study was carried out on Shenzhen and the results showed that: 1) The carbon emission in Shenzhen is unbalanced in spatial and is dense in the west and sparse in the east, but the carbon emission of different networks shows a similar fluctuation trend over time; 2) multi-layer network represents community structure, while communities of “residence-enterprise” network and “residence-park” network are internally closely connected; 3) the carbon emission reduction potential of residential nodes is low in the west and high in the east. The mode of the urban trip and the law of geographical space connection expressed by it were understood by establishing a multi-layer network embedded in geographic space in this paper. The conclusions hereof are of supportive significance for the formulation of space emission reduction policies.

Spatio-temporal analysis of carbon footprints for urban public transport systems based on smart card data

The increasing severity of global climate change has made reductions in carbon emissions an urgent global issue. The relative lack of carbon footprint analyses of urban public transportation systems (UPTS) is therefore surprising, given that UPTS is an important component of urban transportation and one that may play a crucial role in carbon emission reduction. This study conducts a spatio-temporal analysis of carbon footprints for UPTS during the COVID-19 pandemic based on smart card data in Beijing. Since the core of carbon footprint calculation is to estimate travellers' trip trajectories and the ridership of urban rail transit (URT) and buses, we construct a novel multi-layer urban rail network model to calculate passenger volume and travellers' trajectories through a traffic assignment model. Furthermore, we utilize the Generalized Additive Model (GAM) to analyse the correlation relationship between the carbon footprints of buses and URT. Additionally, we conduct statistical analysis of the carbon footprint of UPTS. The results of the spatio-temporal analysis of carbon footprints for UPTS show significantly lower carbon emissions during holidays compared to those on working days, and emissions during peak hours contribute approximately half of the total daily UPTS emissions, while there are notable variations in the distribution of the carbon footprint among different districts. Moreover, our analysis reveals a positive correlation between the carbon footprints of buses and URT. The statistical analysis reflects different patterns of carbon footprint distribution on different dates during the pandemic, but the carbon footprint distributions on selected dates all follow a power-law distribution. This study may facilitate the understanding to the impacts of UPTS on the environment during the COVID-19 pandemic, and also provide important guidance and reference for the development of carbon emission reduction strategies.

Construction and traffic dynamics analysis of multi-layer network model with different speed physical layers

The study of traffic dynamics on multi-layer network contributes to revealing the traffic law of the actual network and suppressing the occurrence of network congestion. Focusing on actual network traffic characteristics, we construct a three-layer network model which consists of one logical layer and two physical layers. The traffic dynamics of the three-layer network model are carefully compared with different logical layer routing strategies. In addition, three indicators of load average, load standard deviation and load reduction rate are proposed to more intuitively observe the load conditions of physical layer nodes. We also study the impact of the degree distribution of physical layer on traffic capacity of three-layer network. It is found that when the preference of packets selection for small degree nodes is insufficient or excessive, it is not conducive to achieving higher traffic capacity.

A modularity total environment network model for sustainable forest transportation

Forest road network planning and development is important in providing accessibility to remote forest areas for timber harvesting, transportation to markets, and recreational activities, as well as preventing environmental degradation, soil erosion, water pollution, and an increased risk of forest fires and wildlife habitat fragmentation. Within the context that careful forest road transportation planning and design promotes sustainable forest management and development, this paper develops a multilayer network model for supporting sustainable forest transportation development. The model builds on a total environment conceptualization and the network paradigm, and it is composed of several layers, each containing information from an environmental aspect or a forest road network land use. By using community detection analysis from network science, the model provides insights into the decomposition of the forest network into functional areas, highlights the importance of places that connect different communities to maintain market integration, and provides a list of policies and good practices for the resulting communities. Overall, this paper presents a quantitative methodological framework that can be used for sustainable forest transportation.

Study on hydroturbine power trend prediction based on machine learning

The application of computational fluid dynamics combined with 3D modeling of the hydraulic model was discussed in this paper. Through the complex calculation of the internal low field, dynamic simulation was used to predict the characteristics of water energy index. Based on the collection of a large number of different types of turbine operation data in different working conditions, the improved multi-layer neural network power prediction model based on the adaptive inverse normalization strategy and the improved grey wolf algorithm were used to optimize the least squares support vector machine power prediction model. Rlue activation function was used in the improved multi-layer neural network model uses to improve the computation speed and prevent gradient disappearance. Combined with L2 regularization to prevent data overfitting, Adam gradient optimizer was applied to achieve fast convergence of network output. The simulation results shown that the method can realize the high precision prediction of turbine output power.

Modeling and monitoring multilayer attributed weighted directed networks via a generative model

As data with network structures are widely seen in diverse applications, the modeling and monitoring of network data have drawn considerable attention in recent years. When individuals in a network have multiple types of interactions, a multilayer network model should be considered to better characterize its behavior. Most existing network models have concentrated on characterizing the topological structure among individuals, and important attributes of individuals are largely disregarded in existing works. In this article, first, we propose a unified static Network Generative Model (static-NGM), which incorporates individual attributes in network topology modeling. The proposed model can be utilized for a general multilayer network with weighted and directed edges. A variational expectation maximization algorithm is developed to estimate model parameters. Second, to characterize the time-dependent property of a network sequence and perform network monitoring, we extend the static-NGM model to a sequential version, namely, the sequential-NGM model, with the Markov assumption. Last, a sequential-NGM chart is developed to detect shifts and identify root causes of shifts in a network sequence. Extensive simulation experiments show that considering attributes improves the parameter estimation accuracy and that the proposed monitoring method also outperforms the three competitive approaches, static-NGM chart, score test-based chart (ST chart) and Bayes factor-based chart (BF chart), in both shift detection and root cause diagnosis. We also perform a case study with Enron E-mail data; the results further validate the proposed method.

Pressure drop for subcooled water boiling in microchannels under high heat fluxes: Experiments and predictions from artificial neural networks

Subcooled flow boiling in microchannels holds considerable promise in compact cooling applications. This study reports experimental measurements of the pressure drop of subcooled flow boiling of water in microchannels with inner diameters ranging from 0.5875 mm to 2 mm. The experiments are conducted over a range of operating parameters, including q = 3.0–10 MW/m2, p = 3.0–5.0 MPa, and G = 2000–10,000 kg/(m2∙s). The influences of diameter, heated length, and operating parameters are comprehensively analyzed. The results indicate that an increase in heat flux and pressure leads to an advancement of the pressure drop under the same thermodynamic vapor quality. Furthermore, a multi-layer artificial neural networks model optimized by a genetic algorithm is established using over 1000 experimental data. This model is employed to predict the subcooled boiling pressure drop and determine the optimal dimensionless parameter combination for developing a new pressure drop empirical correlation. The optimal combination of parameters includes Ja, ρl/ρg, Lsb/Lsat, Co and We, and the new correlation is proposed to calculate subcooled boiling pressure drop for high-heat-flux and microchannels applications, exhibiting a mean absolute error (MAE) of 10.64%. The results provide a novel approach that integrates machine learning techniques with empirical correlation, aiming to meet the requirements of the design optimization for thermal management solutions.

Experimental and numerical investigation of heat transfer and flow of water-based graphene oxide nanofluid in a double pipe heat exchanger using different artificial neural network models

In this study, a water-based graphene oxide nanofluid was utilized in a counter-current flow double pipe heat exchanger (DPHE). The inner pipe carried the hot fluid (deionized water-based fluid). In the outer pipe, the cold fluid was flown, in which the experiments were performed once for deionized water-based fluid and once for graphene oxide nanofluid. These experiments were conducted at various inlet hot fluid temperatures of 35, 45, and 55 °C, volumetric concentrations of 0.01, 0.055, and 0.1% for nanofluid, and cold fluid flow rates of 14.4, 18.9, and 23.4 ml/s to evaluate the thermohydraulics of the DPHE. The results demonstrated that an increase in the flow rate and concentration of nanoparticles led to an improvement in the heat transfer coefficient (HTC) . It is also observed that inlet hot fluid temperature had a negligible effect on this coefficient compared to the concentration and flow rate. Additionally, the results showed that the friction factor and pressure drop of nanofluid were higher than those of the basefluid and their values were increased by increasing the concentration. The maximum increase compared to the basefluid was 72% for the friction factor and 111%, for the pressure drop. Besides, radial basis function neural network (RBFNN) and multi-layer perceptron neural network (MLPNN) models were designed for estimating the HTC. Both models accurately predicted the coefficient, but the RBFNN model exhibited higher accuracy than the MLPNN model. The results further indicated that the nanofluid outperformed the basefluid in terms of HTC. Due to the exceptionally high thermal conductivity of graphene nanoparticles, the HTC of the nanofluid was improved by up to 85% compared to the basefluid. The optimum value of HTC was achieved at a temperature of 55 °C, a volume concentration of 0.1%, and a flow rate of 23.4 ml/s.

Development and external validation of a machine learning model for prediction of survival in undifferentiated pleomorphic sarcoma.

Machine learning (ML) algorithms to predict cancer survival have recently been reported for a number of sarcoma subtypes, but none have investigated undifferentiated pleomorphic sarcoma (UPS). ML is a powerful tool that has the potential to better prognosticate UPS. The Surveillance, Epidemiology, and End Results (SEER) database was queried for cases of histologically confirmed undifferentiated pleomorphic sarcoma (UPS) (n = 665). Patient, tumor, and treatment characteristics were recorded, and ML models were developed to predict 1-, 3-, and 5-year survival. The best performing ML model was externally validated using an institutional cohort of UPS patients (n = 151). All ML models performed best at the 1-year time point and worst at the 5-year time point. On internal validation within the SEER cohort, the best models had c-statistics of 0.67-0.69 at the 5-year time point. The Multi-Layer Perceptron Neural Network (MLP) model was the best performing model and used for external validation. Similarly, the MLP model performed best at 1-year and worst at 5-year on external validation with c-statistics of 0.85 and 0.81, respectively. The MLP model was well calibrated on external validation. The MLP model has been made publicly available at https://rachar.shinyapps.io/ups_app/ . Machine learning models perform well for survival prediction in UPS, though this sarcoma subtype may be more difficult to prognosticate than other subtypes. Future studies are needed to further validate the machine learning approach for UPS prognostication.