Multilayer Network Model Research Articles

Deep Graph Learning-Based Surrogate Model for Inverse Modeling of Fractured Reservoirs

Inverse modeling can estimate uncertain parameters in subsurface reservoirs and provide reliable numerical models for reservoir development and management. The traditional simulation-based inversion method usually requires numerous numerical simulations, which is time-consuming. Recently, deep learning-based surrogate models have been widely studied as an alternative to numerical simulation, which can significantly improve the solving efficiency of inversion. However, for reservoirs with complex fracture distribution, constructing the surrogate model of numerical simulation presents a significant challenge. In this work, we present a deep graph learning-based surrogate model for inverse modeling of fractured reservoirs. Specifically, the proposed surrogate model integrates the graph attention mechanisms to extract features of fracture network in reservoirs. The graph learning can retain the discrete characteristics and structural information of the fracture network. The extracted features are subsequently integrated with a multi-layer recurrent neural network model to predict the production dynamics of wells. A surrogate-based inverse modeling workflow is then developed by combining the surrogate model with the differential evolutionary algorithm. Numerical studies performed on a synthetic naturally fractured reservoir model with multi-scale fractures illustrate the performance of the proposed methods. The results demonstrate that the proposed surrogate model exhibits promising generalization performance of production prediction. Compared with tens of thousands of numerical simulations required by the simulation-based inverse modeling method, the proposed surrogate-based method only requires 1000 to 1500 numerical simulations, and the solution efficiency can be improved by ten times.

Accelerated prediction of stacking fault energy in FCC medium entropy alloys using multilayer perceptron neural networks: correlation and feature analysis

A multilayer perceptron neural networks (MLPNN) model is developed for robust and quick prediction of stacking fault energy (SFE) to overcome the challenges faced in the calculation of SFE via experimentation and atomistic calculations in FCC medium entropy alloys (MEA). The present investigation employs a three-step hybrid feature selection approach to obtain a comprehensive understanding of the prominent features that influence the SFE, as well as the interrelationships among these features. The feature space encompasses various features related to composition, lattice stability, and elemental properties, of MEAs. The findings indicate that the estimation of SFE relies on five crucial factors: temperature, lattice stability, specific heat, ionization energy, and Allen electronegativities. Furthermore, a mathematical relationship for the estimation of the SFE is derived, considering the various influencing and prominent factors. Consequently, the MLPNN model for robust SFE prediction in MEAs is developed and the performance is evaluated using R 2 scores, with values of 0.87 and 0.85 obtained for the training and testing datasets, respectively. This efficient strategy introduces a novel opportunity for the engineering of SFE in the extensive range of alloy chemistry of MEAs, enabling the quick prediction of SFE, and facilitating the systematic exploration of new alloys for the development of mechanisms that may accommodate deformation through octahedral/partial slip, twinning, and/or transformation.

Identification of breast cancer subtypes based on graph convolutional network

Identification of molecular subtypes of malignant tumors plays a vital role in individualized diagnosis, personalized treatment, and prognosis prediction of cancer patients. The continuous improvement of comprehensive tumor genomics database and the ongoing breakthroughs in deep learning technology have driven further advancements in computer-aided tumor classification. Although the existing classification methods based on gene expression omnibus database take the complexity of cancer molecular classification into account, they ignore the internal correlation and synergism of genes. To solve this problem, we propose a multi-layer graph convolutional network model for breast cancer subtype classification combined with hierarchical attention network. This model constructs the graph embedding datasets of patients' genes, and develops a new end-to-end multi-classification model, which can effectively recognize molecular subtypes of breast cancer. A large number of test data prove the good performance of this new model in the classification of breast cancer subtypes. Compared to the original graph convolutional neural networks and two mainstream graph neural network classification algorithms, the new model has remarkable advantages. The accuracy, weight-F1-score, weight-recall, and weight-precision of our model in seven-category classification has reached 0.851 7, 0.823 5, 0.851 7 and 0.793 6 respectively. In the four-category classification, the results are 0.928 5, 0.894 9, 0.928 5 and 0.865 0 respectively. In addition, compared with the latest breast cancer subtype classification algorithms, the method proposed in this paper also achieved the highest classification accuracy. In summary, the model proposed in this paper may serve as an auxiliary diagnostic technology, providing a reliable option for precise classification of breast cancer subtypes in the future and laying the theoretical foundation for computer-aided tumor classification.

Can Machine Learning Predict Running Kinematics Based on Upper Trunk GPS-Based IMU Acceleration? A Novel Method of Conducting Biomechanical Analysis in the Field Using Artificial Neural Networks

This study aimed to investigate whether running kinematics can be accurately estimated through an artificial neural network (ANN) model containing GPS-based accelerometer variables and anthropometric data. Thirteen male participants with extensive running experience completed treadmill running trials at several speeds. Participants wore a GPS device containing a triaxial accelerometer, and running kinematics were captured by an 18-camera motion capture system for each trial. Multiple multilayer perceptron neural network models were constructed to estimate participants’ 3D running kinematics. The models consisted of the following input variables: 3D peak accelerometer acceleration during foot stance (g), stance time (s), running speed (km/h), participant height (cm), leg length (cm), and mass (kg). Pearson’s correlation coefficient (r), root mean squared error (RMSE), and relative root mean squared error (rRMSE) showed that ANN models provide accurate estimations of joint/segment angles (mean rRMSE = 13.0 ± 4.3%) and peak segment velocities (mean rRMSE = 22.1 ± 14.7%) at key gait phases across foot stance. The highest accuracies were achieved for flexion/extension angles of the thorax, pelvis, and hip, and peak thigh flexion/extension and vertical velocities (rRMSE < 10%). The current findings offer sports science and medical practitioners working with this data a method of conducting field-based analyses of running kinematics using a single IMU.

Hybrid MLP-CNN-based ground sink susceptibility prediction in urban area using underground pipe map

In urban areas, ground sinks are generally attributed to underground pipe damages resulting in leakage. Hence, it is essential to forecast ground sinks in advance because they might cause detrimental effects on infrastructure, including the roads, bridges, and railroads. Prior studies have focused on geological and topographical factors, with an inconsiderable amount of attention paid to the study of underground pipes. This study proposes a hybrid multilayer perceptron-convolutional neural network (MLP-CNN) model to predict ground sink susceptibility in urban areas using an underground pipe map. A dataset was constructed by extracting table and image data from the underground pipe map, and it was used to train the model. The test results indicated that the model could accurately predict ground sink susceptibility with an accuracy of 93.87 % and f1-score of 0.9478. Furthermore, we generated a ground sink susceptibility map (GSSM) using a trained model. The findings of this study confirmed that the hybrid MLP-CNN model is capable of reliably predicting ground sink susceptibility.

RatInABox, a toolkit for modelling locomotion and neuronal activity in continuous environments.

Generating synthetic locomotory and neural data is a useful yet cumbersome step commonly required to study theoretical models of the brain's role in spatial navigation. This process can be time consuming and, without a common framework, makes it difficult to reproduce or compare studies which each generate test data in different ways. In response, we present RatInABox, an open-source Python toolkit designed to model realistic rodent locomotion and generate synthetic neural data from spatially modulated cell types. This software provides users with (i) the ability to construct one- or two-dimensional environments with configurable barriers and visual cues, (ii) a physically realistic random motion model fitted to experimental data, (iii) rapid online calculation of neural data for many of the known self-location or velocity selective cell types in the hippocampal formation (including place cells, grid cells, boundary vector cells, head direction cells) and (iv) a framework for constructing custom cell types, multi-layer network models and data- or policy-controlled motion trajectories. The motion and neural models are spatially and temporally continuous as well as topographically sensitive to boundary conditions and walls. We demonstrate that out-of-the-box parameter settings replicate many aspects of rodent foraging behaviour such as velocity statistics and the tendency of rodents to over-explore walls. Numerous tutorial scripts are provided, including examples where RatInABox is used for decoding position from neural data or to solve a navigational reinforcement learning task. We hope this tool will significantly streamline computational research into the brain's role in navigation.

Comparison of Graph Distance Measures for Movie Similarity Using a Multilayer Network Model.

Graph distance measures have emerged as an effective tool for evaluating the similarity or dissimilarity between graphs. Recently, there has been a growing trend in the application of movie networks to analyze and understand movie stories. Previous studies focused on computing the distance between individual characters in narratives and identifying the most important ones. Unlike previous techniques, which often relied on representing movie stories through single-layer networks based on characters or keywords, a new multilayer network model was developed to allow a more comprehensive representation of movie stories, including character, keyword, and location aspects. To assess the similarities among movie stories, we propose a methodology that utilizes a multilayer network model and layer-to-layer distance measures. We aim to quantify the similarity between movie networks by verifying two aspects: (i) regarding many components of the movie story and (ii) quantifying the distance between their corresponding movie networks. We tend to explore how five graph distance measures reveal the similarity between movie stories in two aspects: (i) finding the order of similarity among movies within the same genre, and (ii) classifying movie stories based on genre. We select movies from various genres: sci-fi, horror, romance, and comedy. We extract movie stories from movie scripts regarding character, keyword, and location entities to perform this. Then, we compute the distance between movie networks using different methods, such as the network portrait divergence, the network Laplacian spectra descriptor (NetLSD), the network embedding as matrix factorization (NetMF), the Laplacian spectra, and D-measure. The study shows the effectiveness of different methods for identifying similarities among various genres and classifying movies across different genres. The results suggest that the efficiency of an approach on a specific network type depends on its capacity to capture the inherent network structure of that type. We propose incorporating the approach into movie recommendation systems.

Industrial and textile waste trade: Multilayer network and environmental policy effects

Waste management is an international enterprise, and it is important to understand global flows of recyclable materials. The Pollution Haven Hypothesis (PHH) suggests that waste moves from high income nations with stringent environmental policy to low income nations with less environmentally stringent policy, by exploiting low labor and regulatory costs. This paper assesses the PHH thesis for slag/dross and textiles (SDT) wastes in PHH through novel integration of the multilayer network and gravity models. The multilayer network model generates network effect that quantifies interlayer connections of multiple waste trade networks. Instead of North-South movement of waste, North-North, South-South, and even South-North are shown. Results from the gravity model indicate that stringent waste management policies reduce both waste exports and imports. PHH is not found for slags/dross where high income countries are importing the waste, contradicting PHH. On the other hand, PHH is more evident between highly connected hubs and havens in SDT waste trade networks.

Analyzing predictors of pearl millet supply chain using an artificial neural network

PurposeThe internal (farmer-controlled) and external (non-farmer-controlled) factors affect crop yield. However, not a single study has identified and analyzed yield predictors in India using effective predictive models. Thus, this study aims to investigate how internal and external predictors impact pearl millet yield and Stover yield.Design/methodology/approachDescriptive analytics and artificial neural network are used to investigate the impact of predictors on pearl millet yield and Stover yield. From descriptive analytics, 473 valid responses were collected from semi-arid zone, and the predictors were categorized into internal and external factors. Multi-layer perceptron-neural network (MLP-NN) model was used in Statistical Package for the Social Sciences version 25 to model them.FindingsThe MLP-NN model reveals that rainfall has the highest normalized importance, followed by irrigation frequency, crop rotation frequency, fertilizers type and temperature. The model has an acceptable goodness of fit because the training and testing methods have average root mean square errors of 0.25 and 0.28, respectively. Also, the model has R2 values of 0.863 and 0.704, respectively, for both pearl millet and Stover yield.Research limitations/implicationsTo the best of the authors’ knowledge, the current study is first of its kind related to impact of predictors of both internal and external factors on pearl millet yield and Stover yield.Originality/valueThe literature reveals that most studies have estimated crop yield using limited parameters and forecasting approaches. However, this research will examine the impact of various predictors such as internal and external of both yields. The outcomes of the study will help policymakers in developing strategies for stakeholders. The current work will improve pearl millet yield literature.

A two-layer stacking regression model for predicting yarn quality of small samples

In addressing the prevalent challenges associated with the limited generalization ability and weak adaptability to small samples observed in neural network models in contemporary yarn quality prediction research, this study proposes a two-layer stacking regression model. The initial layer of the stacking model integrates three base models: random forest, gradient boosting decision tree, and adaptive boosting. The secondary layer employs the linear regression algorithm. The efficacy of the stacking model was juxtaposed against the performance assessment of a multilayer perceptron neural network model, in addition to evaluations of the single random forest, gradient boosting decision tree, and adaptive boosting models, respectively. The generalization ability of the models was assessed through experiments that randomly partitioned the sample dataset based on 10 distinct random seeds. The adaptability to small samples of the models was evaluated through experiments spanning nine varying training set sample sizes. The experiment results underscore that the stacking model outperforms in aspects of prediction accuracy, generalization ability, and adaptability to small samples.

Deconstructing cultural appropriation in online communities: A multilayer network analysis approach

In this study, we introduce a novel multilayer network model designed to analyze complex social phenomena in online communities. The model captures intricate relationships between users, content, and specific aspects of social phenomena, providing a comprehensive framework for understanding these interactions. We applied this model to a dataset of over 1 million Reddit comments from January to April 2022, filtered for cultural appropriation-related keywords. Our quantitative analyses, based on Social Network Analysis techniques, revealed significant findings. For instance, a subreddit exhibited the highest user interaction, indicating a substantial level of engagement on this topic. Furthermore, the distribution of key contents across different subreddits was non-uniform, suggesting diverse levels of engagement across communities. The results of this research underscore the potential of our approach in providing a nuanced understanding of social phenomena in online communities, thereby contributing to future research in this field.

NEURAL NETWORK BASED ESTIMATION OF ECOLOGICAL FOOTPRINT IN TURKEY WITH ECONOMIC AND ENVIRONMENTAL FACTORS

The globe is currently confronted with escalating ecological challenges as a result of the exponential growth of the population, the expansion of industrial activities, the rapid urbanization process, and the heightened levels of consumption. Due to the prevailing ecological issues, there has been an unbridled surge in the demand for natural resources. The emergence of the Ecological Footprint was driven by the observable escalation of environmental degradation, with the primary objective of promoting sustainability. In this study, an Artificial Neural Network (ANN) model was developed for the estimation of ecological footprint. In the first stage, the variables affecting the ecological footprint were determined. It was concluded that there are strong relationships between the independent variables of Gross Domestic Product (GDP), KOF Globalization Index (KOFGI), and Natural Resource Rent (NRR) and the dependent variable of ecological footprint. In the light of this information, the data of GDP (% annual increase), NRR (% GDP) and KOF Globalization Index and ecological footprint data of Turkey between 1970 and 2016 were used to present the ANN model. Feed-Forward Backprop Method, which is one of the multi-layer network models, was applied in the modeling of the ANN. Levenberg-Marquardt optimization, which updates the weight and bias values of the network, was used as the network training function. The study's results suggest that there was a strong correlation between the whole dataset and the model's accuracy, with a close match of 99.316%. Based on the findings, it can be deduced that the developed artificial neural network (ANN) model has a significant level of precision in forecasting the Ecological Footprint.

Last-Mile Optimization Using Neural Networks

In the era of extensive data acquisition from manufacturing and transportation processes, the utilization of machine learning and deep learning techniques has emerged as a potent force for informed decision-making and optimized deliveries in contemporary urban landscapes. This study presents a novel approach grounded in deep learning, where product data are systematically gathered to construct a multilayer perceptron neural network model. This model proves instrumental in efficiently classifying product flows within the urban milieu. To validate its efficacy, machine learning classifiers are deployed, and their performance is juxtaposed with the neural network model. Addressing the critical question of the paper’s significance, our experimental evaluation unequivocally demonstrates the superior classification accuracy of the proposed multilayer perceptron model when compared to traditional machine learning models operating on the same product dataset. This advancement is not merely a theoretical achievement but translates into tangible improvements in last-mile delivery processes, marked by significant cost reduction and the mitigation of delays. The transformative potential of our approach is further underscored by the strategic application of a deep learning algorithm for optimization and illustrative purposes. This holistic methodology not only positions our work as a noteworthy contribution to the realm of product classification but also establishes a concrete pathway for enhancing the sustainability and efficiency of urban logistics. This paper, thus, goes beyond the conventional application of machine learning models, offering a paradigm shift in the intersection of deep learning, urban logistics, and sustainable development.

Characterization and optimization of 5´ untranslated region containing poly-adenine tracts in Kluyveromyces marxianus using machine-learning model

BackgroundThe 5´ untranslated region (5´ UTR) plays a key role in regulating translation efficiency and mRNA stability, making it a favored target in genetic engineering and synthetic biology. A common feature found in the 5´ UTR is the poly-adenine (poly(A)) tract. However, the effect of 5´ UTR poly(A) on protein production remains controversial. Machine-learning models are powerful tools for explaining the complex contributions of features, but models incorporating features of 5´ UTR poly(A) are currently lacking. Thus, our goal is to construct such a model, using natural 5´ UTRs from Kluyveromyces marxianus, a promising cell factory for producing heterologous proteins.ResultsWe constructed a mini-library consisting of 207 5´ UTRs harboring poly(A) and 34 5´ UTRs without poly(A) from K. marxianus. The effects of each 5´ UTR on the production of a GFP reporter were evaluated individually in vivo, and the resulting protein abundance spanned an approximately 450-fold range throughout. The data were used to train a multi-layer perceptron neural network (MLP-NN) model that incorporated the length and position of poly(A) as features. The model exhibited good performance in predicting protein abundance (average R2 = 0.7290). The model suggests that the length of poly(A) is negatively correlated with protein production, whereas poly(A) located between 10 and 30 nt upstream of the start codon (AUG) exhibits a weak positive effect on protein abundance. Using the model as guidance, the deletion or reduction of poly(A) upstream of 30 nt preceding AUG tended to improve the production of GFP and a feruloyl esterase. Deletions of poly(A) showed inconsistent effects on mRNA levels, suggesting that poly(A) represses protein production either with or without reducing mRNA levels.ConclusionThe effects of poly(A) on protein production depend on its length and position. Integrating poly(A) features into machine-learning models improves simulation accuracy. Deleting or reducing poly(A) upstream of 30 nt preceding AUG tends to enhance protein production. This optimization strategy can be applied to enhance the yield of K. marxianus and other microbial cell factories.

Combining Li Zehou’s Aesthetic Theory and Emerging Technologies - Exploring the Construction of an Interactive Academic Field

Abstract The use of emerging technologies to construct an interactive academic field of Li Zehou’s aesthetic theory aims to promote the study of Li Zehou’s aesthetic thought and the development of contemporary Chinese aesthetics. Based on the content and characteristics of Li Zehou’s aesthetics theory, the article describes the innovative application of emerging technology in academic communication, constructs the academic field of Li Zehou’s aesthetics theory on the basis of Bourdieu’s field theory, and designs the framework of the interactive academic field by combining with emerging technology. Relying on Li Zehou’s interactive academic field for the construction of a multi-layer innovation network model and providing relevant quantitative methods for interactive academic field network metrics. The dataset is constructed with the relevant literature of Li Zehou’s aesthetics theory from 2005 to 2022 as the search target, and the network density, central potential, clustering coefficients, and path lengths of the interactive field are analyzed, and the evolution characteristics of the multilayer field network are explored. The cross-border innovation multilayer network density of the interactive field of Li Zehou’s aesthetic thought fluctuates around 0.092, approaching the mean value of central potential up to 65.76%, and the average clustering coefficient fluctuates in the range of [0.25,0.35]. The cooperative sub-network and the technology sub-network have smaller path lengths with larger clustering coefficients, indicating small-worldness. The interactive academic field combining emerging technologies can promote the widespread dissemination of Li Zehou’s aesthetic ideas.

Prediction of Skin Lesions Using Integrated Multi-Layered Network Model with Baseline Learning Approaches

Skin cancer has become more common in recent decades, raising severe concerns about world health. Creating an automated system to distinguish between benign and malignant images is challenging because of the subtle variations in how skin lesions appear. This study introduces Computer-Aided Diagnosis (CAD) system that offers high classification accuracy while maintaining low computing complexity for categorizing skin lesions. The system incorporates a pre-processing stage that uses morphological filtering to remove hair and artefacts. With the least minimum of human interaction, deep learning techniques are employed to separate skin lesions automatically. Image processing methods are currently being utilized to investigate the automated implementation of the prediction criteria for distinguishing between benign and malignant melanoma lesions. Various pre-trained convolutional neural networks (CNNs) with multi-layered (ML-CNN) are under examination for the classification of skin lesions as either benign or malignant. The best performance is achieved when RF, k-NN and XGBoost are combined, according to average 5-fold cross-validation findings. The outcomes also demonstrate that data augmentation works better than acquiring novel images for training and testing purposes. The experiment results show that the suggested diagnostic framework performs better than existing methods when used on actual clinical skin lesions, with accuracy at 97.5%, F1-score at 91.3%, precision at 96.5%, sensitivity at 89.2% and specificity at 96.7%. It also takes 2.6 seconds to complete with the MNIST dataset and accuracy at 98.2%, F1-score at 92.5%, precision at 98.4%, sensitivity at 92.3% and specificity of 97.2% with the ISIC dataset. This indicates that medical professionals can benefit from using the suggested framework to classify various skin lesions.

Advancing nanofluid technology: Experimental thermal conductivity measurements and accurate predictive models for yttrium oxide-water nanofluid using neural networks and mathematical correlations

This work involved the experimental measurement of thermal conductivity (TC) values for a novel nanofluid. This nanofluid was created by incorporating yttrium oxide nanoparticles into pure water, which served as the base fluid. The TC values of the nanofluid consisting of yttrium oxide and water were empirically determined at five distinct concentrations. These measurements were conducted within the temperature range of 10–65 degrees Celsius. The data obtained from the experiment demonstrated a 1.47% enhancement in the TC of the nanofluid consisting of yttrium oxide and water, in comparison to the base fluid. Experimental results were compared with correlations commonly used in the literature. A multilayer neural network model and a novel mathematical correlation were built utilizing a dataset of 60 experimental observations to ascertain the relationship between temperature, concentration, and the TC of yttrium oxide-water nanofluid. Nine data sets were used for validation of the network model, and in addition, the model was validated with various performance parameters. The values derived from the network model and mathematical correlation were juxtaposed with both the empirical findings and the outcomes documented in existing literature. The R value for the network model was calculated as 0.99980 and the MSE value was 5.89E-07. The average deviation rate for the network model was determined to be −0.0007%, whereas the mathematical correlation yielded a value of 0.049%. The study’s results demonstrated the development of a neural network model and a novel mathematical connection for accurately calculating the TC of yttrium oxide-water nanofluid.

Geochemical evaluation and source rock zonation by multi-layer perceptron neural network technique: a case study for Pabdeh and Gurpi Formations-North Dezful Embayment (SW Iran)

In this study, using a multi-layer perceptron neural network (MLPNN) model, total organic carbon (TOC) and hydrogen index (HI) values for Pabdeh and Gurpi Formations in the oil fields of Naft Sefid (NS-13), Kupal (KL-36, KL-38, and KL-48) and Palangan (PL-2) were calculated in the North Dezful Embayment located in the southwest of Iran. To build the MLPNN model, the geochemical data calculated by the Rock–Eval pyrolysis method (TOC and HI) and the conventional petrophysical well log data, including sonic transit time log (DT), formation density log (RHOB), total resistivity log (RT), spectral gamma-ray log, computed gamma-ray log and neutron porosity log from the NS-13 well were used. The log data were the input layer, and the geochemical data were the output layer of the model. Twenty-four datasets were used for MLPNN training, and seven datasets were used for MLPNN testing. Two hidden layers were considered in this technique. Each hidden layer has an activation function (tanh) and a solver parameter (lbfgs). The accuracy of measurement of TOC and HI indices of Pabdeh and Gurpi Formations in terms of R2 was 0.93 and 0.90, respectively. This model has higher accuracy than the ΔlogR technique (R2: 0.28). Considering the relationships between the input data and other wireline logs is an advantage of this technique. These two formations have five source rock zones. Pabdeh Formation has three zones. The middle zone of the Pabdeh Formation (Pz. II) has the highest TOC (2.6 wt%) and source rock potential. Pabdeh Formation has kerogen type II. Gurpi Formation has a weaker source rock potential than Pabdeh Formation due to its low TOC content (< 1%). Both source rock zones of this formation have low TOC, but in some layers of the lower zone of the Gurpi Formation (Gz. II), high values for TOC were predicted. Gurpi Formation has Kerogen types II and III.

Spatio-temporal prediction of land use and land cover change in Bahi (Manyoni) Catchment, Tanzania, using multilayer perceptron neural network and cellular automata-Markov chain model.

Evaluation of land use and land cover (LULC) change is among vital tools used for tracking environmental health and proper resource management. Remote sensing data was used to determine LULC change in Bahi (Manyoni) Catchment (BMC) in central Tanzania. Landsat satellite images from Landsat 5 TM and Landsat 8 OLI/TIRS were used, and support vector machine (SVM) algorithm was applied to classify the features of BMC. The obtained kappa values were 0.74, 0.83 and 0.84 for LULC maps of 1985, 2005 and 2021, respectively, which indicates the degree of accuracy from produced being substantial to almost perfect. Classified maps along with geospatial, socio-economic and climatic drivers with sufficient explanatory power were incorporated into MLP-NN to produce transition potential maps. Transition maps were subsequently used in cellular automata (CA)-Markov chain model to predict future LULC for BMC in immediate-future (2035), mid-future (2055) and far-future (2085). The findings indicate BMC is expected to experience significant expansion of agricultural lands and built land from 31.89 to 50.16% and 1.48 to 9.1% from 2021 to 2085 at the expense of open woodland, shrubland and savanna grassland. Low-yield crop production, water scarcity and population growth were major driving forces for rapid expansion of agricultural lands and overall LULC in BMC. The findings are essential for understanding the impact of LULC on hydrological processes and offer insights for the internal drainage basin (IDB) board to make necessary measures to lessen the expected dramatic changes in LULC in the future while sustaining harmonious balance with livelihood activities.

Using Deep-Learning for 5G End-to-End Delay Estimation Based on Gaussian Mixture Models

Deep learning is used in various applications due to its advantages over traditional Machine Learning (ML) approaches in tasks encompassing complex pattern learning, automatic feature extraction, scalability, adaptability, and performance in general. This paper proposes an end-to-end (E2E) delay estimation method for 5G networks through deep learning (DL) techniques based on Gaussian Mixture Models (GMM). In the first step, the components of a GMM are estimated through the Expectation-Maximization (EM) algorithm and are subsequently used as labeled data in a supervised deep learning stage. A multi-layer neural network model is trained using the labeled data and assuming different numbers of E2E delay observations for each training sample. The accuracy and computation time of the proposed deep learning estimator based on the Gaussian Mixture Model (DLEGMM) are evaluated for different 5G network scenarios. The simulation results show that the DLEGMM outperforms the GMM method based on the EM algorithm, in terms of the accuracy of the E2E delay estimates, although requiring a higher computation time. The estimation method is characterized for different 5G scenarios, and when compared to GMM, DLEGMM reduces the mean squared error (MSE) obtained with GMM between 1.7 to 2.6 times.