Multilayer Network Model Research Articles

Applying the Stackelberg game to assess critical infrastructure vulnerability: Based on a general multi-layer network model.

Critical infrastructure systems (CIS) are closely related to human life. Attacks against CIS occur frequently, making accurate and effective protection of CIS essential. Vulnerability assessment is the primary issue to achieving this goal. The interconnected characteristic of CIS means that it is best represented by a multi-layer network, but a uniform model is absent. Game theory offers a suitable framework for researching intelligent confrontation. Previous research combining game theory and network science mainly focuses on a single-layer network and lacks a comprehensive assessment that combines qualitative and quantitative aspects of vulnerability. In this paper, we apply the Stackelberg game to the multi-layer network and comprehensively assess vulnerability based on game equilibrium to realize accurate protection. We first present a method for constructing a general model of the multi-layer network and introduce a multi-layer weighted factor to extend topological attribute metrics. Then, we design a Stackelberg game model for the multi-layer network. Furthermore, we qualitatively analyze the impact of multi-layer network characteristics on vulnerability and propose a method to quantify vulnerability. Experiments show that the vulnerability of the multi-layer network is greatly influenced by the multi-layer weighted factor, single-layer network type, and inter-layer coupling method. The quantitative value of network vulnerability does not rely entirely on the topology attributes on which the attack and defense costs depend but also relates to the resources available. Our work provides an adaptive model for CIS and gives a new approach to developing accurate protection based on comprehensive vulnerability assessment, which deserves further study.

Natural and Artificial Intelligence Interactions in Digital Networking: A Multilayer Network Model for Economic Value Creation

This study investigates the integration of natural intelligence (NI) and artificial intelligence (AI) within traditional firms linked through a multilayer network framework. The research explores this central question: How can the integration of NI and AI, facilitated by copula nodes, drive economic value creation in digitized firms? The paper combines theoretical and empirical results of the observed increments in cost-benefit marginality linked with AI adoption by checking out testing in various domains such as manufacturing, retailing, finance, etc. The performance of the model shows enhanced efficiency and decision-making, as well as savings in costs. Copula nodes that connect multilayer networks blend NI with AI, boosting overall value. Unlike existing studies, this framework operationalizes copula nodes to capture real-time dependencies between human-driven and AI-driven processes, offering a comprehensive view of economic value creation across digitized firms. A methodology section outlines the empirical validation process conducted across multiple industries (manufacturing, retail, and finance), providing key insights into efficiency, decision-making, and cost optimization. The practical implications offer a strategic pathway for firms to enhance profitability and competitiveness in the digital age. The results underscore the importance of strategically integrating AI with human intelligence to maximize economic outcomes. Received: 18 August 2024 | Revised: 22 October 2024 | Accepted: 10 November 2024 Conflicts of Interest The author declares that he has no conflicts of interest to this work. Data Availability Statement Data sharing is not applicable to this article as no new data were created or analyzed in this study. Author Contribution Statement Roberto Moro Visconti: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data Curation, Writing - Original Draft, Writing -Review & Editing, Visualization, Supervision, Project administration, Funding acquisition.

Assessment of mass transfer performance using the two-film theory and surrogate models for intensified CO2 capture process by amine solutions in rotating packed beds

CO2 capture is a crucial aspect of attempts to mitigate climate change. The purpose of this study is to investigate the impacts of essential operating parameters on the mass transfer performance of absorbing CO2 in rotating packed beds (RPBs). A multilayer perceptron neural network (MLPNN) model with the Levenberg-Marquardt learning algorithm, a mass transfer model using the two-film theory, and an empirical correlation model were developed to predict the overall gas-phase volumetric mass-transfer coefficient (KGaV) for RPB-based CO2 absorption. The developed MLPNN model showed excellent agreement with the actual data, with an MSE of 0.0357, an AARD of 7.4%, and an R2 of 0.9839. A sensitivity analysis was conducted using Taguchi orthogonal array design on distinct mass transfer correlations. The results of the two-film theory and surrogate models for the diethylenetriamine (DETA) solvent were compared. The MLPNN model provided better predictions than other developed models with an AARD of 13% for CO2H2O-DETA system. Therefore, the effects of operating parameters such as concentration, temperature, solvent flow rate, and rotational speed on KGaVand CO2 removal efficiency were evaluated using the MLPNN model. Finally, an empirical correlation was proposed to predict KGaVas a function of operational parameters.

Explainable predictions of a machine learning model to forecast the postoperative length of stay for severe patients: machine learning model development and evaluation.

Predicting the length of stay in advance will not only benefit the hospitals both clinically and financially but enable healthcare providers to better decision-making for improved quality of care. More importantly, understanding the length of stay of severe patients who require general anesthesia is key to enhancing health outcomes. Here, we aim to discover how machine learning can support resource allocation management and decision-making resulting from the length of stay prediction. A retrospective cohort study was conducted from January 2018 to October 2020. A total cohort of 240,000 patients' medical records was collected. The data were collected exclusively for preoperative variables to accurately analyze the predictive factors impacting the postoperative length of stay. The main outcome of this study is an analysis of the length of stay (in days) after surgery until discharge. The prediction was performed with ridge regression, random forest, XGBoost, and multi-layer perceptron neural network models. The XGBoost resulted in the best performance with an average error within 3days. Moreover, we explain each feature's contribution over the XGBoost model and further display distinct predictors affecting the overall prediction outcome at the patient level. The risk factors that most importantly contributed to the stay after surgery were as follows: a direct bilirubin laboratory test, department change, calcium chloride medication, gender, and diagnosis with the removal of other organs. Our results suggest that healthcare providers take into account the risk factors such as the laboratory blood test, distributing patients, and the medication prescribed prior to the surgery. We successfully predicted the length of stay after surgery and provide explainable models with supporting analyses. In summary, we demonstrate the interpretation with the XGBoost model presenting insights on preoperative features and defining higher risk predictors to the length of stay outcome. Our development in explainable models supports the current in-depth knowledge for the future length of stay prediction on electronic medical records that aids the decision-making and facilitation of the operation department.

Optimizing resource allocation for enhanced urban connectivity in LEO-UAV-RIS networks

Sixth-generation (6G) communication advancements target massive connectivity, ultra-reliable low-latency communication (URLLC), and high data rates, essential for IoT applications. Yet, in natural disasters, particularly in dense urban areas, 6G quality of service (QoS) can falter when terrestrial networks—such as base stations—become unavailable, unstable, or strained by high user density and dynamic environments. Additionally, high-rise buildings in smart cities contribute to signal blockages. To ensure reliable, high-quality connectivity, integrating low-Earth Orbit (LEO) satellites, unmanned aerial vehicles (UAVs), and reconfigurable intelligent surfaces (RIS) into a multilayer (ML) network offers a solution: LEO satellites provide broad coverage, UAVs reduce congestion with flexible positioning, and RIS enhances signal quality. Despite these benefits, this integration brings challenges in resource allocation, requiring path loss models that account for both line-of-sight (LOS) and non-line-of-sight (NLOS) links. To address these, a joint optimization problem is formulated focusing on resource distribution fairness. Given its complexity, a framework is proposed to decouple the problem into subproblems using the block coordinate descent (BCD) method. These subproblems include UAV placement optimization, user association, subcarrier allocation via orthogonal frequency division multiple access (OFDMA), power allocation, and RIS phase shift control. OFDMA efficiently manages shared resources and mitigates interference. This iterative approach optimizes each subproblem, ensuring convergence to a locally optimal solution. Additionally, we propose a low-complexity solution for RIS phase shift control, proving its feasibility and efficiency mathematically. The numerical results demonstrate that the proposed scheme achieves up to 43.5% higher sum rates and 80% lower outage probabilities compared to the schemes without RIS. The low complexity solution for RIS optimization achieves performance within 1.8% of the SDP approach in terms of the sum rate. This model significantly improves network performance and reliability, achieving a 16.3% higher sum rate and a 44.4% reduction in outage probability compared to joint optimization of SAT-UAV resources. These findings highlight the robustness and efficiency of the ML network model, making it ideal for next-generation communication systems in high-density urban environments.

Integrative analysis of single-cell transcriptomic and multilayer signaling networks in glioma reveal tumor progression stage.

Tumor microenvironments (TMEs) encompass complex ecosystems of cancer cells, infiltrating immune cells, and diverse cell types. Intercellular and intracellular signals within the TME significantly influence cancer progression and therapeutic outcomes. Although computational tools are available to study TME interactions, explicitly modeling tumor progression across different cancer types remains a challenge. This study introduces a comprehensive framework utilizing single-cell RNA sequencing (scRNA-seq) data within a multilayer network model, designed to investigate molecular changes across glioma progression stages. The heterogeneous, multilayered network model replicates the hierarchical structure of biological systems, from genetic building blocks to cellular functions and phenotypic manifestations. Applying this framework to glioma scRNA-seq data allowed complex network analysis of different cancer stages, revealing significant ligand‒receptor interactions and key ligand‒receptor-transcription factor (TF) axes, along with their associated biological pathways. Differential network analysis between grade III and grade IV glioma highlighted the most critical nodes and edges involved in interaction rewiring. Pathway enrichment analysis identified four essential genes-PDGFA (ligand), PDGFRA (receptor), CREB1 (TF), and PLAT (target gene)-involved in the Receptor Tyrosine Kinases (RTK) signaling pathway, which plays a pivotal role in glioma progression from grade III to grade IV. These genes emerged as significant features for machine learning in predicting glioma progression stages, achieving 87% accuracy and 93% AUC in a 3-year survival prediction through Kaplan-Meier analysis. This framework provides deeper insights into the cellular machinery of glioma, revealing key molecular relationships that may inform prognosis and therapeutic strategies.

Advancing Geotechnical Evaluation of Wellbores: A Robust and Precise Model for Predicting Uniaxial Compressive Strength (UCS) of Rocks in Oil and Gas Wells

This study examines the efficacy of various machine learning models for predicting the uniaxial compressive strength (UCS) of rocks in oil and gas wells, which are essential for ensuring wellbore stability and optimizing drilling operations. The investigation encompasses Linear Regression, ensemble methods (including Random Forest, Gradient Boosting, XGBoost, and LightGBM), support vector machine-based regression (SVM-SVR), and multilayer perceptron artificial neural network (MLP-ANN) models. The results demonstrate that XGBoost and Gradient Boosting offer superior predictive accuracy for UCS in drillability, as indicated by low Mean Absolute Percentage Error (MAPE) values of 3.87% and 4.18%, respectively, and high R2 scores (0.8542 for XGBoost). These models emerge as optimal choices for UCS prediction focused on drillability, offering increased accuracy and reliability in practical engineering scenarios. Ensemble methods and MLP-ANN emerge as frontrunners, providing valuable tools for improving wellbore stability assessments, optimizing drilling parameter selection, and facilitating informed decision-making processes in oil and gas drilling operations. Moreover, this study lays a foundation for further research in drillability-centred predictive modelling for geotechnical parameters, advancing our understanding of rock behaviour under drilling conditions.

Behavioral Modeling of Drivers near Speed Control Cameras: A Dual Perspective from Micro and Macro Data

The present study investigates the efficacy of speed cameras in managing vehicle speeds and their influence on various driver demographics within the operational context of the cameras. The research, conducted as a case study in Iran, comprises both macroscopic and detailed analyses, quantifying vehicle speeds at different spatial points relative to the speed camera location—upstream, adjacent to, and downstream of the camera—using a manual laser speedometer. Furthermore, the study examines diverse driver behavior metrics, including vehicle type, speeds at different locations, lane changes, braking actions, and measures such as obscuring license plates. Statistical analyses show that speed cameras effectively reduce speeds across different user groups and lead to increased braking and lane-changing behaviors, particularly among light vehicle categories. Moreover, the study reveals distinct behavioral patterns between indigenous and non-indigenous drivers. In addition, a multi-layer perceptron neural network model successfully approximates user speed selection behaviors within the operational range of speed cameras. Overall, this research provides valuable insights into the effectiveness of speed cameras and their impact on diverse driver demographics, contributing to a deeper understanding of advanced speed control mechanisms.

Medical image recognition based on multilayer neural network

As a common accidental injury, accurate identification of burns and scalded injuries is of great significance to the development of treatment programs and prognosis assessment. Traditional identification methods are subjective and inaccurate, this paper proposes a burns and scalds grade identification technique based on multilayer neural network. The three-degree classification standard of burns and scalds is described, and the structure and principle of multilayer neural network are introduced, including the network structure composed of input layer, hidden layer and output layer, forward propagation, back propagation and training process. The implementation steps of burns and scald grade recognition technology based on multi-layer neural network are discussed in detail, such as data acquisition, image pre-processing, feature extraction, classifier design, model training and evaluation. Through experiments, the model is tested using a dataset containing 362 burns and scalds images, and the model achieves more than 90% accuracy on the test set with high accuracy and reliability. The multi-layer neural network model is used to classify the burn and scald data, which improves the diagnostic accuracy, shortens the delay time of the disease, reduces the pressure of professional doctors, and helps patients have a preliminary understanding of the degree of burn.

A modified deep learning method for Alzheimer’s disease detection based on the facial submicroscopic features in mice

Alzheimer’s disease (AD) is a chronic disease among people aged 65 and older. As the aging population continues to grow at a rapid pace, AD has emerged as a pressing public health issue globally. Early detection of the disease is important, because increasing evidence has illustrated that early diagnosis holds the key to effective treatment of AD. In this work, we developed and refined a multi-layer cyclic Residual convolutional neural network model, specifically tailored to identify AD-related submicroscopic characteristics in the facial images of mice. Our experiments involved classifying the mice into two distinct groups: a normal control group and an AD group. Compared with the other deep learning models, the proposed model achieved a better detection performance in the dataset of the mouse experiment. The accuracy, sensitivity, specificity and precision for AD identification with our proposed model were as high as 99.78%, 100%, 99.65% and 99.44%, respectively. Moreover, the heat maps of AD correlation in the facial images of the mice were acquired with the class activation mapping algorithm. It was proven that the facial images contained AD-related submicroscopic features. Consequently, through our mouse experiments, we validated the feasibility and accuracy of utilizing a facial image-based deep learning model for AD identification. Therefore, the present study suggests the potential of using facial images for AD detection in humans through deep learning-based methods.

Photometric redshift estimation for CSST survey with LSTM neural networks

ABSTRACT Accurate estimation of photometric redshifts (photo-zs) is crucial for cosmological surveys. Various methods have been developed for this purpose, such as template fitting methods and machine learning techniques, each with its own applications, advantages, and limitations. In this study, we propose a new approach that utilizes a deep learning model based on recurrent neural networks with long short-term memory (LSTM) to predict photo-z. Unlike many existing machine learning models, our method requires only flux measurements from different observed filters as input. The model can automatically learn the complex relationships between the flux data across different wavelengths, eliminating the need for manually extracted or derived input features, thereby providing precise photo-z estimates. The effectiveness of our proposed model is evaluated using simulated data from the Chinese Space Station Telescope (CSST) sourced from the Hubble Space Telescope Advanced Camera for Surveys (HST-ACS) and the COSMOS catalogue, considering anticipated instrument effects of the future CSST. Results from experiments demonstrate that our LSTM model, compared to commonly used template fitting and machine learning approaches, requires minimal input parameters and achieves high precision in photo-z estimation. For instance, when trained on the same data set and provided only with photometric fluxes as input features, the proposed LSTM model yields one-third of the outliers $f_{\mathrm{out}}$ observed with a multi-layer perceptron neural network (MLP) model, while the normalized median absolute deviation $\rm \sigma _{NMAD}$ is only two-thirds that of the MLP model. This study presents a novel approach to accurately estimate photo-zs of galaxies using photometric data from large-scale survey projects.

Landscape and environmental heterogeneity support coexistence in competitive metacommunities

Metapopulation models have been instrumental in quantifying the ecological impact of landscape structure on the survival of a focal species. However, extensions to multiple species with arbitrary dispersal networks often rely on phenomenological assumptions that inevitably limit their scope. Here, we propose a multilayer network model of competitive dispersing metacommunities to investigate how spatially structured environments impact species coexistence and ecosystem stability. We introduce the concept of landscape-mediated fitness, quantifying how fit a species is in a given environment in terms of colonization and extinction. We show that, when all environments are equivalent, one species excludes all the others-except the marginal case where species fitnesses are in exact trade-off. However, we prove that stable coexistence becomes possible in sufficiently heterogeneous environments by introducing spatial disorder in the model and solving it exactly in the mean-field limit. Crucially, coexistence is supported by the spontaneous localization of species through the emergence of ecological niches. We show that our results remain qualitatively valid in arbitrary dispersal networks, where topological features can improve species coexistence by buffering competition. Finally, we employ our model to study how correlated heterogeneity promotes spatial ecological patterns in realistic terrestrial and riverine landscapes. Our work provides a framework to understand how landscape structure enables coexistence in metacommunities by acting as the substrate for ecological interactions.

Predicting biohydrogen production from dark fermentation of organic waste biomass using multilayer perceptron artificial neural network (MLP–ANN)

The focus on sustainable energy has increased interest in biohydrogen production through dark fermentation of organic waste biomass, offering dual benefits of energy production and waste management. Optimizing this process is challenging due to complex interactions among substrate composition, microbial consortia, and fermentation parameters. A multilayer perceptron artificial neural network model was developed to predict biohydrogen yield from organic waste. The model, trained on 180 data points from 35 studies, uses inputs, such as substrate type, inoculum type, concentration, pH, and temperature, with hydrogen yield as the output. The multilayer perceptron artificial neural network model achieved high accuracy, with a root mean square error of 0.3838, a mean absolute percentage error of 0.1938, and a coefficient of determination of 0.8381. These results demonstrate the model's effectiveness in predicting biohydrogen production, providing a valuable tool for optimizing the fermentation process and advancing sustainable energy solutions.

Machine learning to classify left ventricular hypertrophy using ECG feature extraction by variational autoencoder.

Traditional ECG criteria for left ventricular hypertrophy (LVH) have low diagnostic yield. Machine learning (ML) can improve ECG classification. ECG summary features (rate, intervals, axis), R-wave, S-wave and overall-QRS amplitudes, and QRS/QRST voltage-time integrals (VTIs) were extracted from 12-lead, vectorcardiographic X-Y-Z-lead, and root-mean-square (3D) representative-beat ECGs. Latent features were extracted by variational autoencoder from X-Y-Z and 3D representative-beat ECGs. Logistic regression, random forest, light gradient boosted machine (LGBM), residual network (ResNet) and multilayer perceptron network (MLP) models using ECG features and sex, and a convolutional neural network (CNN) using ECG signals, were trained to predict LVH (left ventricular mass indexed in women >95 g/m², men >115 g/m²) on 225,333 adult ECG-echocardiogram (within 45 days) pairs. AUROCs for LVH classification were obtained in a separate test set for individual ECG variables, traditional criteria and ML models. In the test set (n=25,263), AUROC for LVH classification was higher for ML models using ECG features (LGBM 0.790, MLP 0.789, ResNet 0.788) as compared to the best individual variable (VTIQRS-3D 0.677), the best traditional criterion (Cornell voltage-duration product 0.647) and CNN using ECG signal (0.767). Among patients without LVH who had a follow-up echocardiogram >1 (closest to 5) years later, LGBM false positives, compared to true negatives, had a 2.63 (95% CI 2.01, 3.45)-fold higher risk for developing LVH (p<0.0001). ML models are superior to traditional ECG criteria to classify-and predict future-LVH. Models trained on extracted ECG features, including variational autoencoder latent variables, outperformed CNN directly trained on ECG signal.

Reconciliation of weak pairwise spike-train correlations and highly coherent local field potentials across space.

Multi-electrode arrays covering several square millimeters of neural tissue provide simultaneous access to population signals such as extracellular potentials and spiking activity of one hundred or more individual neurons. The interpretation of the recorded data calls for multiscale computational models with corresponding spatial dimensions and signal predictions. Multi-layer spiking neuron network models of local cortical circuits covering about $1\,{\text{mm}^{2}}$ have been developed, integrating experimentally obtained neuron-type-specific connectivity data and reproducing features of observed in-vivo spiking statistics. Local field potentials can be computed from the simulated spiking activity. We here extend a local network and local field potential model to an area of $4\times 4\,{\text{mm}^{2}}$, preserving the neuron density and introducing distance-dependent connection probabilities and conduction delays. We find that the upscaling procedure preserves the overall spiking statistics of the original model and reproduces asynchronous irregular spiking across populations and weak pairwise spike-train correlations in agreement with experimental recordings from sensory cortex. Also compatible with experimental observations, the correlation of local field potential signals is strong and decays over a distance of several hundred micrometers. Enhanced spatial coherence in the low-gamma band around $50\,\text{Hz}$ may explain the recent report of an apparent band-pass filter effect in the spatial reach of the local field potential.

A method for assessing the resilience of urban interdependent systems integrating physical damage and social loss

As systems in urban areas become increasingly interdependent, the effects of disasters extend beyond facilities damage, precipitating casualties, and public service disruptions. Establishing a holistic resilience assessment method is crucial for urban sustainability planning. Existing resilience assessment methods often focus on the physical damage; however, few studies examine the social loss. This study proposes a resilience assessment method that integrates physical damage and social loss. A multilayer network model of physical-social systems is first developed based on their interdependency. Then, the evolution of disaster damage within and between systems process, and the multi-task parallel restoration process are quantified. Resilience is assessed by measuring dynamic changes in service supply and demand throughout these processes. The method considers social loss as a combination of population harm and service interruption caused by physical damage. The proposed method was applied to an urban area in Southern China. The results reveal the impact of seismic events on infrastructure, buildings, and populations. Furthermore, they demonstrate the coupling impacts of supply and transportation on emergency and healthcare systems’ resilience. This comprehensive method could facilitate the development of efficient strategies to improve urban resilience to disasters.

Aspect based sentiment analysis of consumer reviews using unsupervised attention neural framework

Consumer reviews posted on e-commerce platforms offer an assessment of their opinion and sentiments about various aspects of a product or service. The aspects may belong to a specific category at subsystem and component level. The irrealis often associated with an opinion term, if not handled carefully, may distort the sentiment scores. This paper introduces a clustering and negative-sampling backed feed-forward Unsupervised Attention Deep Neural Framework for Opinion-Aspect-Category-Irrealis-Sentiment (UAN-OACIS) Quintuple extraction. This framework splits the five subtasks into two stages. In the first stage, aspects and opinions are extracted using word embedding-based multi-layer Attention Deep Neural Network (DNN) models; while the second stage projects corresponding categories, irrealis, and sentiment scores by creating a hash table and a couple of algorithms. We also propose a POS-Tag-based algorithm to make an irrealis dictionary to suggest sentiment modification percentage scores; and another algorithm to modify existing sentiment-lexicon based on domain-specific words. We compare UAN-OACIS with four unsupervised and two semi-supervised approaches at subtask level, and two supervised approaches at quadruples level excluding irrealis, and have got comparable results. Since no method is available for overall comparison at quintuple level, we additionally study the impact of irrealis on sentiment scores and observe substantial improvement compared to state-of-the-art lexicons methods.

The Forecasting of TheNumber Tourists Arriving in Turkey with Intuitionistic Fuzzy Regression Functions Approach

The impacts of the tourism sector on countries are felt in various areas such as economy, cultural heritage and social development. Tourism contributes significantly to a country's foreign exchange earnings and positively affects the trade network. Tourists' spending boosts local economies and increases employment. These effects are particularly important for Turkey. Tourist visits can be used as a tool for regional promotion. Therefore, tourism demand forecasting is necessary to make the best use of these positive effects on Turkey's economic development and to plan tourism activities. Artificial neural network methods and fuzzy systems for time series forecasting problem are frequently used analysis methods in recent years. In this study, the time series of the total number of tourists visiting Turkey on a monthly basis is analyzed with the intuitionistic fuzzy regression functions approach, which is a generalization of the fuzzy regression functions approach. The analysis performance of the intuitionistic fuzzy regression functions approach is evaluated using fuzzy regression functions approach, multilayer perceptron artificial neural network and multiplicative neuron model artificial neural networks. As a result of the analysis, it is concluded that the intuitionistic fuzzy regression approach produces better forecasting results than both some artificial neural network models and the fuzzy regression functions approach. Since this is the first time that the intuitionistic fuzzy regression functions approach has been used in forecasting the number of tourists, the study aims to contribute to the literature and to help tourism industry employees to be more efficient and successful by providing them with the opportunity to make better future planning.

Enhancing solar energy conversion efficiency: Thermophysical property predicting of MXene/Graphene hybrid nanofluids via bayesian-optimized artificial neural networks

Accurately predicting thermo-physical properties (TPPs) of MXene/graphene-based nanofluids is crucial for photovoltaic/thermal solar systems, driving focused research on developing precise TPP predictive models. This study presents optimized multi-layer perceptron neural network (MLPNN) models, leveraging Bayesian optimization to refine architectural and training hyperparameters, including hidden layers, neurons, activation functions, standardization, and regularization terms. A comparative analysis of Bayesian acquisition functions—the probability of improvement (POI), lower confidence bound (LCB), expected improvement (EI), expected improvement plus (EIP), expected improvement per second plus (EIPSP), and expected improvement per second (EIPS)—demonstrated that the POI-MLPNN achieves the most accurate results, as evidenced by the lowest MAPE of 1.0923 % and exceptional consistency with an R-value of 0.99811. The EI-MLPNN and EIP-MLPNN models recorded the same outputs. The EI/EIP-MLPNN (R = 0.99668) model excels in consistency over LCB-MLPNN (R = 0.99529) and EIPSP-MLPNN (R = 0.99667). The optimized models offer a reliable, cost-efficient alternate for experimental and computational TPP analyses. Leveraging insights from these models enables better control over nanofluid TPPs in solar systems, enhancing energy conversion efficiency.

Prediction Model using a Multi-Layer Perceptron Neural Network for Military Plastic Explosive PE4 Blast Performance in a Significant Effect of Tropicalisation

Over the years, plastic explosive PE4 has been imported from the United Kingdom and is fully utilised in military activities and drills. This study is to discover a ratio of the plastic explosive PE4 explosion performance in tropical conditions and develop a multilayer perceptron neural network model for the prediction of plastic explosive in Malaysia. Several environmental tests have been performed in Kem Kongkoi, Jelebu, Negeri Sembilan. Six parameters were considered, including environmental temperature, distance measurement, explosive material, variety of shapes, weight and ignition point. In this paper, the Bayesian Regularization, Levenberg-Marquardt algorithm and Scaled Conjugate Gradient model were developed. Every model was tested with 24 datasets to discover the root mean square error and regression performance. The Bayesian Regularization model provides the best prediction model as it has a mean square error of 0.0005 and a regression performance value that is close to 1 at 0.9992.