Neural Network Based Prediction Model Research Articles

Artificial Neural Network-Based Model for Prediction of Frost Heave Behavior of Silty Soil Specimen

Frost heave action is a major issue in permafrost regions that can give rise to various geotechnical engineering problems. To analyze and predict this phenomenon at a specimen scale, this study conducted a fully coupled thermal-hydro-mechanical analysis and evaluated the frost heave behavior of frozen soil considering geotechnical parameters. Furthermore, a parametric study was performed to quantitatively analyze the effects of major geotechnical properties on frost heave behavior. According to the results of the parametric study, the amount of heave tended to decrease as the particle thermal conductivity increased, whereas the frost heave ratio tended to increase as the initial hydraulic conductivity increased. After evaluating the sensitivity of each parameter to frost heave behavior through statistical analyses, an artificial neural network model was developed to practically predict frost heave behavior. According to the verification results of the neural network model, the trained network model demonstrated a reliable accuracy (R2 = 0.893) in predicting frost heave ratio, even when the model used test datasets that were not part of the training datasets.

An efficient Capacitor Bank Operating System for Single Phase Power Factor Correction using Neural Network Estimations

Wastage of electricity occurs in all places starting from a small house electrical loading to a heavy industrial electrical loading. KiloVolt-Ampere Reactive (KVAR) power metering devices are employed in industrial applications for measuring the energy utilization which measure the energy wastage along with it. This urges a consumer to pay for the unutilized or wasted energy as well. To avoid this, certain capacitor bank units are connected to the industrial application motor units. The right choice of capacitor rating are helpful in minimizing the wasted power observation in the KVAR meters. The selection of capacitor rating is analysed with respect to the power factor calculation. The power factor is a derivation of working power to the apparent power in an electrical system. An optimum power factor to be maintained in an electrical system is 1. The motive of the proposed work is to maintain the power factor by selecting an optimum capacitor bank on the operation of an electrical system at various load conditions. The requirement of capacitor bank values get changed with respect to the load given to an electrical system. A neural network based prediction model is employed in the work for estimating the right choice of capacitor bank. The efficiency of the proposed work is verified and found satisfied with a traditional capacitor bank operating system.

Forecasting the demand for tourism using combinations of forecasts by neural network-based interval grey prediction models

ABSTRACT In contrast to point forecasting, interval forecasting provides the degree of variation associated with forecasts. Accurate forecasting can help governments formulate policies for tourism, but little attention has been paid to interval forecasting of tourism demand. This study contributes to apply neural networks to develop interval models for tourism demand forecasting. Since combined forecasts are likely to improve the accuracy of point forecasting, forecast combinations are used to construct the proposed models. Besides, grey prediction models without requiring that data follow any statistical assumption serve as constituent models. Empirical results show that the proposed models outperform other considered interval models.

WITHDRAWN: Artificial neural network-based psychological assessment model for predicting the mental health problem in children facing psychological abuse and depression

Risk assessment is critical to prevent psychological abuse of children as it can classify high-risk situations requiring action to protect children. Despite the widespread use of risk evaluation tools in child care, the prediction of mental health issues and the properties of the tools are consistent with increased predictive efficacy are still unclear. The psychological abuse of children is associated with an increased risk of depression. However, depression is present in many people who receive abuse. Hence in this paper, Artificial Neural Network-based Psychological Symptom Prediction Model (ANN-PSM) has been proposed to reduce depression and improve children's psychological level. The ANN-PSM patterns can detect the children increasingly likely to create psychological abuse in the face of sad signals and an increased risk of depression. Findings help to explain the mechanisms by which psychiatric depression puts children at risk for socioemotional adaptation. The experimental findings indicate that ANN's can be accepted more generally in clinical decision-making and achieve the highest detection rate of 97.84% depression among children.

Classification of Influenza Hemagglutinin Protein Sequences using Convolutional Neural Networks.

The Influenza virus can be considered as one of the most severe viruses that can infect multiple species with often fatal consequences to the hosts. The Hemagglutinin (HA) gene of the virus can be a target for antiviral drug development realised through accurate identification of its sub-types and possible the targeted hosts. This paper focuses on accurately predicting if an Influenza type A virus can infect specific hosts, and more specifically, Human, Avian and Swine hosts, using only the protein sequence of the HA gene. In more detail, we propose encoding the protein sequences into numerical signals using the Hydrophobicity Index and subsequently utilising a Convolutional Neural Network-based predictive model. The Influenza HA protein sequences used in the proposed work are obtained from the Influenza Research Database (IRD). Specifically, complete and unique HA protein sequences were used for avian, human and swine hosts. The data obtained for this work was 17999 human-host proteins, 17667 avian-host proteins and 9278 swine-host proteins. Given this set of collected proteins, the proposed method yields as much as 10% higher accuracy for an individual class (namely, Avian) and 5% higher overall accuracy than in an earlier study. It is also observed that the accuracy for each class in this work is more balanced than what was presented in this earlier study. As the results show, the proposed model can distinguish HA protein sequences with high accuracy whenever the virus under investigation can infect Human, Avian or Swine hosts.

Ensemble Models of Cutting-Edge Deep Neural Networks for Blood Glucose Prediction in Patients with Diabetes.

This article proposes two ensemble neural network-based models for blood glucose prediction at three different prediction horizons—30, 60, and 120 min—and compares their performance with ten recently proposed neural networks. The twelve models’ performances are evaluated under the same OhioT1DM Dataset, preprocessing workflow, and tools at the three prediction horizons using the most common metrics in blood glucose prediction, and we rank the best-performing ones using three methods devised for the statistical comparison of the performance of multiple algorithms: scmamp, model confidence set, and superior predictive ability. Our analysis provides a comparison of the state-of-the-art neural networks for blood glucose prediction, estimating the model’s error, highlighting those with the highest probability of being the best predictors, and providing a guide for their use in clinical practice.

T-SNE and variational auto-encoder with a bi-LSTM neural network-based model for prediction of gas concentration in a sealed-off area of underground coal mines

A deep learning network is introduced to predict concentrations of gases in the underground coal mine enclosed region using various IoT-enabled gas sensors installed in a metallic gas chamber. The air is sucked automatically at specific intervals from the sealed-off site utilizing a solenoid valve, suction pump, and programmed microprocessor. The gas sensors monitor the gas content in the underground coal mine and communicate gas concentration to the surface server room through a wireless network and cloud storage media. The t-SNE_VAE_bi-LSTM model is proposed in this study as a prediction model that combines the t-SNE, VAE, and bi-LSTM networks. The proposed model's t-SNE method aims to minimize the dimensionality of the recorded gas concentration; and VAE layer intends to retrieve the inner characteristics of low-dimensional gas concentration. Finally, the given model's Bi-LSTM layer tries to forecast the concentrations of CH4, CO2, CO, O2, and H2 gases. The proposed model's prediction accuracy is compared with the existing two models, namely auto-regressive integrated average moving (ARIMA) and chaos time series (CHAOS). The experiment findings demonstrate that the t-SNE_VAE_bi-LSTM model forecasted mean square error (MSE) is more accurate, and it has lesser MSE value of 0.029 and 0.069 for CH4; 0.037 and 0.019 for CO2; 0.092 and 0.92 for CO; 1.881 and 1.892 for O2; and 1.235 and 1.200 for H2 than the ARIMA and CHAOS models, respectively.

Dynamic prediction of penetration rate based on TBM operational data

In this paper, an artificial neural network-based dynamic prediction model of penetration rate (PR) is proposed. Four tunnel boring machine (TBM) operational parameters, including cutterhead rotational speed (RPM), cutterhead torque (T), total thrust (F), and advance rate (AR) are introduced as input of the network. Their data in the trial excavation phase, are employed to predict the PR in the stable excavation phase. Therefore, the training data will have enough offsets from the test data, and the TBM operators will have sufficient time to either fine-tune the operational parameters or shut off the machine if undesired prediction results come out. To examine the performance of the established model, it is applied to a 20 km water conveyance tunnel. The results show that the network converges fast and steadily with acceptable performance on the test set. The major contribution of this paper is verifying the possibility to estimate the PR based on the historical data of the TBM.

NEURAL NETWORK-BASED PREDICTION MODEL 
 FOR THE DYNAMICS OF THE IONOSPHERIC EQUATORIAL ANOMALY
 USING THE TOTAL ELECTRONIC CONTENT

In this paper we present the prediction model for the dynamics of the ionospheric equatorial anomaly that is based on the use of the Principal Component Analysis (PCA) and Artificial Neural Networks (ANN). The prediction model was developed by using global maps of the ionosphere Total Electronic Content (TEC) for the period from 2001 to 2018. We show that in case of correct data centering and elimination of diurnal and seasonal factors, the equatorial anomaly makes major contribution to the variance of fluctuations in the TEC data. We applied several neural network-based prediction models that were trained independently for each component of the decomposition. The approach based on a hybrid model consisting of a convolution network and a network with long short-term memory with preanalysis of the principal components reduced the prediction error of TEC maps by 2 hours. The prediction error of this model was 4 times less than the error of the linear regression model.

Using machine learning to study protein-protein interactions: From the uromodulin polymer to egg zona pellucida filaments.

Neural network-based models for protein structure prediction have recently reached near-experimental accuracy and are fast becoming a powerful tool in the arsenal of biologists. As suggested by initial studies using RoseTTAFold or the ColabFold implementation of AlphaFold2, a particularly interesting future development will be the optimization of these computational methods to also routinely yield high-confidence predictions of protein-protein interactions. Here I use AlphaFold2 and ColabFold to investigate the activation and polymerization of uromodulin (UMOD)/Tamm-Horsfall protein, a zona pellucida (ZP) module-containing protein whose precursor and filamentous structures have been previously determined experimentally by X-ray crystallography and cryo-EM, respectively. Despite having no knowledge of the UMOD polymer structure (coordinates for which were neither used for model training nor as template), AlphaFold2/ColabFold are able to recapitulate a crucial conformational change underlying UMOD polymerization, as well as the general organization of protein subunits within the resulting filament. This surprising result is achieved by simply deleting from the input sequence a stretch of residues that correspond to a polymerization-inhibiting C-terminal propeptide. By mimicking in silico the activating effect of propeptide dissociation triggered by site-specific proteolysis of the protein precursor, this example has implications for the assembly of egg coat proteins and the many other molecules that also contain a ZP module. Most importantly, it shows the potential of exploiting machine learning not only to accurately predict the structures of individual proteins or complexes, but also to carry out computational experiments replicating specific molecular events.

The performance and optimization of ANN-WP model under unknown sea states

Ocean wave is one of the main factors to cause the 6 degrees of freedom motion of ships. Accurate wave surface prediction is paramount to ship safety. The artificial neural network-based wave prediction (ANN-WP) model is a phase-resolved wave prediction model based on artificial intelligence, which can provide high-precision prediction results of wave surface. However, it needs training with a large dataset to make the model obtain a satisfying prediction performance. When unknown sea states (not included in the training data) appear in practice, the reliability of the prediction results is low. In this study, the performance of the ANN-WP model under unknown sea states is investigated based on wave tank experiment. The data distribution of training data set is considered as one of the factors that affects the performance of ANN-WP model under unknown sea states. Based on this consideration, data mixing is employed to improve the performance of the ANN-WP model under unknown sea states. Through this method, the prediction error of the model for unknown sea states decreased by approximately 5.0%–10.0%.

Data-Driven State Prediction and Sensor Fault Diagnosis for Multi-Agent Systems with Application to a Twin Rotational Inverted Pendulum

When a multi-agent system is subjected to faults, it is necessary to detect and classify the faults in time. This paper is motivated to propose a data-driven state prediction and sensor fault classification technique. Firstly, neural network-based state prediction model is trained through historical input and output data of the system. Then, the trained model is implemented to the real-time system to predict the system state and output in absence of fault. By comparing the predicted healthy output and the measured output, which can be abnormal in case of sensor faults, a residual signal can be generated. When a sensor fault occurs, the residual signal exceeds the threshold, a fault classification technique is triggered to distinguish fault types. Finally, the designed data-driven state prediction and fault classification algorithms are verified through a twin rotational inverted pendulum system with leader-follower mechanism.

Basketball Sports Injury Prediction Model Based on the Grey Theory Neural Network.

Sports injuries will have an impact on the consistency and systemicity of the training process, as well as athlete training and performance improvement. Many talented athletes have had their careers cut short due to sports injuries. Preventing sports injuries is the best way for basketball players to reduce sports injuries. Many coaches and athletes on sports teams, on the other hand, are unaware of the importance of sports injury prevention. They only realize that the body's sports functions are abnormal when it suffers from sports injuries. As a result, this paper proposes a gray theory neural network-based athlete injury prediction model. First, from the standpoint of a single model, the improved unequal interval model is used to predict sports injury by optimizing the unequal interval model in gray theory. The findings show that it is a good predictor of sports injuries, but it is a poor predictor of the average number of injuries. Following that, in order to overcome the shortcomings of a single model, a gray neural network combination model was used. A combination model of the unequal time interval model and BP neural network was determined and established. The prediction effect is significantly improved by combining the gray neural network mapping model and the coupling model to predict the two characteristics of sports injuries. Finally, simulation experiments show that the proposed method is effective.

A Deep Neural Network-Based Multi-Frequency Path Loss Prediction Model from 0.8 GHz to 70 GHz.

Large-scale fading models play an important role in estimating radio coverage, optimizing base station deployments and characterizing the radio environment to quantify the performance of wireless networks. In recent times, multi-frequency path loss models are attracting much interest due to their expected support for both sub-6 GHz and higher frequency bands in future wireless networks. Traditionally, linear multi-frequency path loss models like the ABG model have been considered, however such models lack accuracy. The path loss model based on a deep learning approach is an alternative method to traditional linear path loss models to overcome the time-consuming path loss parameters predictions based on the large dataset at new frequencies and new scenarios. In this paper, we proposed a feed-forward deep neural network (DNN) model to predict path loss of 13 different frequencies from 0.8 GHz to 70 GHz simultaneously in an urban and suburban environment in a non-line-of-sight (NLOS) scenario. We investigated a broad range of possible values for hyperparameters to search for the best set of ones to obtain the optimal architecture of the proposed DNN model. The results show that the proposed DNN-based path loss model improved mean square error (MSE) by about 6 dB and achieved higher prediction accuracy R2 compared to the multi-frequency ABG path loss model. The paper applies the XGBoost algorithm to evaluate the importance of the features for the proposed model and the related impact on the path loss prediction. In addition, the effect of hyperparameters, including activation function, number of hidden neurons in each layer, optimization algorithm, regularization factor, batch size, learning rate, and momentum, on the performance of the proposed model in terms of prediction error and prediction accuracy are also investigated.

A Deep Recurrent Neural Network-Based Explainable Prediction Model for Progression from Atrophic Gastritis to Gastric Cancer

Gastric cancer is the fifth most common cancer type worldwide and one of the most frequently diagnosed cancers in South Korea. In this study, we propose DeepPrevention, which comprises a prediction module to predict the possibility of progression from atrophic gastritis to gastric cancer and an explanation module to identify risk factors for progression from atrophic gastritis to gastric cancer, to identify patients with atrophic gastritis who are at high risk of gastric cancer. The data set used in this study was South Korea National Health Insurance Service (NHIS) medical checkup data for atrophic gastritis patients from 2002 to 2013. Our experimental results showed that the most influential predictors of gastric cancer development were sex, smoking duration, and current smoking status. In addition, we found that the average age of gastric cancer diagnosis in a group of high-risk patients was 57, and income, BMI, regular exercise, and the number of endoscopic screenings did not show any significant difference between groups. At the individual level, we identified that there were relatively strong associations between gastric cancer and smoking duration and smoking status.

Attention-based interpretable neural network for building cooling load prediction

Machine learning has gained increasing popularity in building energy management due to its powerful capability and flexibility in model development as well as the rich data available in modern buildings. While machine learning is becoming more powerful, the models developed, especially artificial neural networks like Recurrent Neural Networks (RNN), are becoming more complex, resulting in “darker models” with lower model interpretability. The sophisticated inference mechanism behind machine learning prevents ordinary building professionals from understanding the models, thereby lowering trust in the predictions made. To address this, attention mechanisms have been widely implemented to improve the interpretability of deep learning; these mechanisms enable a deep learning-based model to track how different inputs influence outputs at each step of inference.This paper proposes a novel neural network architecture with an attention mechanism for developing RNN-based building energy prediction, and investigates the effectiveness of this attention mechanism in improving the interpretability of RNN models developed for 24-hour ahead building cooling load prediction. To better understand, explain and evaluate these neural network-based building energy prediction models, the obtained attention vectors (or metric) are used to visualize the influence of different parts of model inputs on the prediction result. This helps the users to understand why predictions are made by the model, as well as how input sequences proportionally influence the output sequences. Further analysis of attention vectors can provide interesting temporal information for understanding building thermal dynamics, like the thermal inertia of the building. The proposed attention-based architecture can be implemented in developing optimal operation control strategies and improving demand and supply management. The model developed based on this architecture is assessed using real building operational data, and shows improved accuracy and interpretability over baseline models (without adopting attention mechanisms). The research results help to bridge the gap between building professionals and advanced machine learning techniques. The insights obtained can be used as guidance for the development, fine-tuning, explanation and debugging of data-driven building energy prediction models.

Neural Network-Based Prediction Model to Investigate the Influence of Temperature and Moisture on Vibration Characteristics of Skew Laminated Composite Sandwich Plates.

The present study deals with the development of a prediction model to investigate the impact of temperature and moisture on the vibration response of a skew laminated composite sandwich (LCS) plate using the artificial neural network (ANN) technique. Firstly, a finite element model is generated to incorporate the hygro-elastic and thermo-elastic characteristics of the LCS plate using first-order shear deformation theory (FSDT). Graphite-epoxy composite laminates are used as the face sheets, and DYAD606 viscoelastic material is used as the core material. Non-linear strain-displacement relations are used to generate the initial stiffness matrix in order to represent the stiffness generated from the uniformly varying temperature and moisture concentrations. The mechanical stiffness matrix is derived using linear strain-displacement associations. Then the results obtained from the numerical model are used to train the ANN. About 11,520 data points were collected from the numerical analysis and were used to train the network using the Levenberg–Marquardt algorithm. The developed ANN model is used to study the influence of various process parameters on the frequency response of the system, and the outcomes are compared with the results obtained from the numerical model. Several numerical examples are presented and conferred to comprehend the influence of temperature and moisture on the LCS plates.

Validating the impact of accounting disclosures on stock market: A deep neural network approach

Firms disclose information either voluntarily or due to the regulator's mandatory requirements, and such disclosures form good sources to know the prospects of a firm. Information in the disclosures and analysts' opinions influence investor-trading behavior, and consequently, affects the asset prices. As sentiments factored in disclosures are a source of market action, this study aims to capture the sentiments from disclosure information to assess asset prices' impact. The paper adopts a deep neural network-based prediction model for conducting sentiment analysis on heterogeneous datasets. We construct a sentiment simulation model of voluntary disclosures to know whether the managers can use the market sentiment as a strategic input to boost market performance by suitably drafting the tone and content of disclosures without compromising their quality and veracity. The Deep Neural Networks with LSTM algorithm is found to outperform the Deep Neural Networks with RNN and other baseline machine learning classifiers in terms of predictive accuracy of the NSE NIFTY50. The variable importance computed also validates that market news, combined with historical indicators, predicts the stock market trend closer to the actual trend.

Prediction of local current distribution in polymer electrolyte membrane fuel cell with artificial neural network

In the development process of a fuel cell, understanding the local current distribution is essentially required to achieve better performance and durability. Therefore, many developers apply a segmented fuel cell to observe current distribution under various operating conditions. With the application, experimental data is collected. This study suggests a utilization method for this collected data to develop a local current prediction model. The details of this neural network-based prediction model are introduced, including the pretreatment of the data. In the pretreatment process, current residual values are used for better prediction performance. As a result, the model predicted local current values with a 2.98% error. With the model, the effects of pressure, temperature, cathode relative humidity, and cathode flow rate on local current distribution trends are analyzed. Since the non-uniform current distribution of a fuel cell often leads to low performances or fast local degradation, the optimal operating condition to achieve current uniformity is acquired with an additional model. This model is developed by switching inputs and outputs of the local current prediction model. With the model application, the uniform current distribution is achieved with a standard deviation of 0.039 A/cm2 under the current load at 1 Acm−2.

Predictors of 30-Day Unplanned Readmission After Carotid Artery Stenting Using Artificial Intelligence.

IntroductionThis study aimed to describe the rates and causes of unplanned readmissions within 30 days following carotid artery stenting (CAS) and to use artificial intelligence machine learning analysis for creating a prediction model for short-term readmissions. The prediction of unplanned readmissions after index CAS remains challenging. There is a need to leverage deep machine learning algorithms in order to develop robust prediction tools for early readmissions.MethodsPatients undergoing inpatient CAS during the year 2017 in the US Nationwide Readmission Database (NRD) were evaluated for the rates, predictors, and costs of unplanned 30-day readmission. Logistic regression, support vector machine (SVM), deep neural network (DNN), random forest, and decision tree models were evaluated to generate a robust prediction model.ResultsWe identified 16,745 patients who underwent CAS, of whom 7.4% were readmitted within 30 days. Depression [p < 0.001, OR 1.461 (95% CI 1.231–1.735)], heart failure [p < 0.001, OR 1.619 (95% CI 1.363–1.922)], cancer [p < 0.001, OR 1.631 (95% CI 1.286–2.068)], in-hospital bleeding [p = 0.039, OR 1.641 (95% CI 1.026–2.626)], and coagulation disorders [p = 0.007, OR 1.412 (95% CI 1.100–1.813)] were the strongest predictors of readmission. The artificial intelligence machine learning DNN prediction model has a C-statistic value of 0.79 (validation 0.73) in predicting the patients who might have all-cause unplanned readmission within 30 days of the index CAS discharge.ConclusionsMachine learning derived models may effectively identify high-risk patients for intervention strategies that may reduce unplanned readmissions post carotid artery stenting.Central IllustrationFigure 2: ROC and AUPRC analysis of DNN prediction model with other classification models on 30-day readmission data for CAS subjectsSupplementary InformationThe online version contains supplementary material available at 10.1007/s12325-021-01709-7.