Group Method Of Data Handling Neural Network Research Articles

Quality evaluation of the forecasting system for the Polish Timescale by means of the Group Method of Data Handling type neural network

The article presents the results aimed at assessing the effectiveness of the applied forecasting system based on the forecasting procedure using a GMDH (Group Method of Data Handling) neural network for the Polish Timescale (UTC(PL)) implemented on the basis of hydrogen maser (HM). The article also aims to show that the application of the forecasting system to implement the UTC(PL) national timescale makes it possible to achieve a quality of this timescale similar to the best of the best timescales. The results of forecasting using the GMDH neural network for UTC(PL) are presented for two prepared time series (TS1 and TS2), which are additionally compared with the UTC - UTC(k) values for this scale. Very good forecasting quality has been achieved for the analysed cases 5 and 6, both for data prepared according to the TS1 and TS2 time series. This is confirmed by the calculated values of forecast quality measures. The obtained research results show that the developed system allows ensuring the accuracy of the UTC(PL) at the level of best timescales. The presented research results will help to convince some of NMIs (National Metrology Institutes), which are not equipped with caesium fountains, to adopt the UTC(k) steering system.

Estimating the compressive strength of lightweight foamed concrete using different machine learning-based symbolic regression techniques

The development of concrete with excellent water and frost resistance providing high level of sound and thermal insulation has triggered the formulation of foamed concrete. However, multiple laboratory studies are required to produce reasonable data to design the relevant codes and mathematics with which design of mixes is made easier at low cost. In this research paper, the artificial intelligence (AI)-based symbolic regression technique estimation of the compressive strength of foamed concrete has been reported. Foamed concrete has been a subject of serious research in sustainable built-environment due to its lightweight and structural functionality. In this research work, data gathering method was applied to gather a globally representative data base comprising concrete density to water density (concrete density g/cm3) (γ/γw), water-cement ratio (W/C), and sand-cement ratio (S/C) as input variable and the compressive strength (Fc) as the study output. The dimensionless factors have been derived to eliminate data handling complexities and improve model performances. The 230 data entries from foamed concrete mixes were partitioned into 75% and 25% for training and validation data, respectively. At the end of the model execution, it was found that the response surface methodology (RSM) produced a symbolic closed-form equation like the genetic programming (GP), evolutionary polynomial regression (EPR), and the group method of data-handling-neural network (GMDH-NN). Even though the RSM closed with a minimum error, the GP, EPR and GMDH-NN were faster in runtime. The overall outcomes show that the GP outclassed the EPR, RSM and the GMDH-NN, though with minor margin. Meanwhile the EPR produced the highest outliers from the ±25% test of accuracy envelope. Overall, the present models outperformed those reported in the literature due the parameter reduction through dimensionless factors derivation and provided a decisive model to predict the Fc of foamed concrete.

Optimization of thermophysical properties of nanofluids using a hybrid procedure based on machine learning, multi-objective optimization, and multi-criteria decision-making

The rheological and thermal behavior of nanofluids in real-world scenarios is significantly affected by their thermophysical properties (TPPs). Therefore, optimizing TPPs can remarkably improve the performance of nanofluids. In this regard, in the present study, a hybrid strategy is proposed that combines machine learning (ML), multi-objective optimization (MOO), and multi-criteria decision-making (MCDM) to select optimal parameters for water-based multi-walled carbon nanotubes (MWCNTs)-oxide hybrid nanofluids. In the first step, four critical TPPs, including density ratio (DR), viscosity ratio (VR), specific heat capacity ratio (SHCR), and thermal conductivity ratio (TCR), are modeled using two efficient ML techniques, the group method of data handling neural network (GMDH-NN) and combinatorial (COMBI) algorithm. In the next step, the superior models are subjected to a four-objective optimization by the well-known non-dominated sorting genetic algorithm II (NSGA-II), which aims to minimize DR/VR and maximize SHCR/TCR. This study considers volume fraction (VF), oxide nanoparticle (NP) type, and system temperature as optimization variables. In the final step, two prominent MCDM techniques, TOPSIS and VIKOR, were used to identify the desirable optimal points from the Pareto fronts generated by the MOO algorithm. ML results reveal the COMBI algorithm's superior reliability in accurately modeling various TPPs. The pattern of Pareto fronts for all oxide-NPs indicated that over one-third of the optimal points have a VF > 1.5 %. On the other hand, the distribution of optimal points across different temperature ranges varied significantly depending on the type of oxide-NPs. For Al2O3-based nanofluid, around 90 % of the optimal points were within 40–50 °C. Conversely, for nanofluids containing CeO2 NPs, only approximately 24 % of the optimal points were found within the same temperature range. Considering diverse scenarios for weighting TPPs in the MCDM process implied that combining CeO2/ZnO oxide-NPs with MWCNTs in water-based nanofluids is highly effective across various real-world applications.

Predicting default risk bancassurance using GMDH and dce-GMDH neural network models

Purpose This study aims to develop a novel approach for predicting default risk in bancassurance, which plays a crucial role in the relationship between interest rates in banks and premium rates in insurance companies. The proposed method aims to improve default risk predictions and assist with client segmentation in the banking system. Design/methodology/approach This research introduces the group method of data handling (GMDH) technique and a diversified classifier ensemble based on GMDH (dce-GMDH) for predicting default risk. The data set comprises information from 30,000 credit card clients of a large bank in Taiwan, with the output variable being a dummy variable distinguishing between default risk (0) and non-default risk (1), whereas the input variables comprise 23 distinct features characterizing each customer. Findings The results of this study show promising outcomes, highlighting the usefulness of the proposed technique for bancassurance and client segmentation. Remarkably, the dce-GMDH model consistently outperforms the conventional GMDH model, demonstrating its superiority in predicting default risk based on various error criteria. Originality/value This study presents a unique approach to predicting default risk in bancassurance by using the GMDH and dce-GMDH neural network models. The proposed method offers a valuable contribution to the field by showcasing improved accuracy and enhanced applicability within the banking sector, offering valuable insights and potential avenues for further exploration.

Introducing effective parameters for predicting job burnout using a self-organizing method based on group method of data handling neural network.

In addition to affecting people's bodily and mental health, the Covid-19 epidemic has also altered the emotional and mental well-being of many workers. Especially in the realm of institutions and privately held enterprises, which encountered a plethora of constraints due to the peculiar circumstances of the epidemic. It was thus anticipated that the present study would use a group method of data handling (GMDH) neural network for analyzing the relationship of demographic factors, Coronavirus, resilience, and the burnout in startups. The test methodology was quantitative. The research examined 384 startup directors and representatives, which is a sizable proportion of the limitless community. The BRCS, the MBI-GS, and custom-made assessments of stress due to the Coronavirus were all used to collect data. Cronbach's alpha confirmed the polls' dependability, and an expert panel confirmed the surveys' authenticity. The GMDH neural network's inherent potential for self-organization was used to choose the most useful properties automatically. The trained network has a three-layered topology with 4, 3, and 2 neurons in each of the hidden layers. The GMDH network has significantly reduced the computational load by using just 7 parameters of marital status, stress of covid-19, job experience, professional efficiency, gender, age, and resilience for burnout categorization. After comparing the neural network's output with the acquired data, it was determined that the constructed network accurately classified all of the information. Among the achievements of this research, high accuracy in predicting job burnout, checking the performance of neural network in determining job burnout and introducing effective characteristics in determination of this parameter can be mentioned.

Selection of effective combination of time and frequency features using PSO-based technique for monitoring oil pipelines

Pipeline installation is a time-consuming and expensive process in the oil sector. Because of this, a pipe is often utilized to carry diverse petroleum products; hence, it is crucial to use a precise and dependable control system to identify the kind and quantity of oil products being transported. This study attempts to identify four petroleum products by using an X-ray tube-based system, feature extraction in the frequency and temporal domains, and feature selection using Particle Swarm Optimization (PSO) in conjunction with a Group Method of Data Handling (GMDH) neural network. A sodium iodide detector, a test pipe that simulates petroleum compounds, and an X-ray source make up the implemented system. The detector's output signals were transmitted to the frequency domain, where the amplitudes of the top five dominant frequencies could be determined. Furthermore, the received signals were analyzed to extract five temporal characteristics-MSR, 4th order moment, skewness, WL, and kurtosis. The PSO system takes into account the extracted time and frequency features as input in order to introduce the optimal combination. Four different GMDH neural networks were constructed, and the chosen characteristics were used as inputs for those networks. Finding the volume ratio of each product was the responsibility of each neural network. The four designed neural networks were able to predict the amount of ethylene glycol, crude oil, gasoil, and gasoline with RMSE of 0.26, 0.17, 0.19, and 0.23, respectively. One compelling argument for using the proposed approach in the oil industry is that it can calculate the volume ratio of products with a root mean square error of no more than 0.26. The adoption of a feature selection method to choose the best ones is credited with this remarkable degree of precision. By providing appropriate inputs to neural networks, the control system has significantly outperformed its predecessors in terms of precision and efficiency.

Predicting the returns of the US real estate investment trust market: evidence from the group method of data handling neural network

PurposeThe Group Method of Data Handling (GMDH) neural network has demonstrated good performance in data mining, prediction, and optimization. Scholars have used it to forecast stock and real estate investment trust (REIT) returns in some countries and region, but not in the United States (US) REIT market. The primary goal of this study is to predict the US REIT market using GMDH and then compare its accuracy with that derived from the traditional prediction method.Design/methodology/approachTo forecast the return on the US REIT index, this study used the GMDH neural network and the generalized autoregressive conditional heteroscedasticity (GARCH) model. In this test, the training samples, testing samples, and kernel functions of the GMDH model are controlled to investigate their impact on the accuracy of the machine learning approach. Corresponding experiments were performed using the GARCH model, and the accuracies of these two approaches were compared.FindingsCompared with GARCH, GMDH’s accuracy is much higher, indicating that the machine learning approach can provide a highly accurate prediction of REIT prices. The size of the training samples and the kernel functions in the GMDH model affect the accuracy of the prediction results. In particular, the kernel function has a significant impact on prediction accuracy. The linear and linear covariance kernel functions are simple to train and yield accurate predictions, whereas the quadratic function is difficult to train. Even with small training samples, GMDH can outperform GARCH in prediction accuracy.Research limitations/implicationsAlthough GMDH shows good performance in predicting the US REIT return, it is still a black-box model, and the algorithm is difficult for financial analysts to develop and customize. The data used in this study come from the US REIT market, which is the world’s largest and most liquid market.Social implicationsThis research shows that the GMDH model outperforms the GARCH model in forecasting REIT returns. Hence, investors can use the machine learning approach to make more accurate predictions of the target REITs’ returns and thus better investment decisions. Future investors and researchers may use GMDH to forecast the performance of REITs in other markets.Originality/valueThis is the first study to apply the GMDH neural network to the US REIT market and determine the impact of the two factors on its performance. For example, this research first discusses the impact of kernel functions on the US REIT market using the GMDH neural network. It also includes short-term daily prediction returns that were not previously considered, making it a valuable reference for financial industry analysts.

A comprehensive investigation on the influences of optimal CNC wood machining variables on surface quality and process time using GMDH neural network and bees optimization algorithm

The present research work is concerned with the effects of machining parameters on the quality and economical aspects of wooden product. Process variables include depth of cut, feed rate, spindle speed, and step over in three levels and the purpose of the current study is to minimize the surface roughness and the process time, simultaneously. Each variable has three levels and as a result, the number of experimental tests in full factorial mode was considered to 81. Group method of data handling (GMDH) has been employed to train the neural network and extract the relationships governing the machining process. Optimization of the machining parameters has been done using the bees algorithm (BA). The outcomes obtained from GMDH-BA hybrid technique demonstrated that the combined objective function of the surface roughness and process time has been improved about 23% in comparison with the best experiment in all practical tests. The value of surface roughness and the process time in the optimized experimental test were reduced to 4.36 µm and 93 s, respectively.

Forecasting returns volatility of cryptocurrency by applying various deep learning algorithms

The study aims at forecasting the return volatility of the cryptocurrencies using several machine learning algorithms, like neural network autoregressive (NNETAR), cubic smoothing spline (CSS), and group method of data handling neural network (GMDH-NN) algorithm. The data used in this study is spanning from April 14, 2017, to October 30, 2020, covering 1296 observations. We predict the volatility of four cryptocurrencies, namely Bitcoin, Ethereum, XRP, and Tether, and compare their predictive power in terms of forecasting accuracy. The predictive capabilities of CSS, NNETAR, and GMDH-NN are compared and evaluated by mean absolute error (MAE) and root-mean-square error (RMSE). Regarding the return volatility of Bitcoin and XRP markets, the forecasted results remarkably suggest that in contrast to rival approaches, the CSS can be an effective model to boost the predicting accuracy in the sense that it has the lowest forecast errors. Considering the Ethereum markets’ volatility, the MAE and RMSE associated with NNETAR are smaller than the MAE and RMSE of CSS and GMDH-NN algorithm, which ensures the effectiveness of NNETAR as compared to competing approaches. Similarly, in case of Tether markets’ volatility, the corresponding MAE and RMSE reveal that the GMDH-NN algorithm is an efficient technique to enhance the forecasting performance. We notice that no single tool performed uniformly for all cryptocurrency markets. The policymakers can adopt the model for forecasting cryptocurrency volatility accordingly.

A miniaturized filter design approach using GMDH neural networks

AbstractA miniaturized microstrip low‐pass filter (LPF) has been designed and fabricated using semicircular‐shaped resonators. The designing procedures as well as the LC equivalent circuit of each step have been reported. Due to the small size of the presented filter, flat group delay in the pass‐band, low insertion loss (IL), and high return loss (RL) in operational frequency. By producing adjusted transmission zeros (TZs), the second to fourth unwanted harmonics were attenuated using the proposed LPF. The machine learning approach has been utilized to adjustment of the filter parameters. The 3 dB cut‐off frequency (Fc), TZs, IL, and RL were predicted using the group method of data handling neural network (GMDHNN). Ninety simulations were performed with different inductor and capacitor values. The collected data were used for neural network training. The GMDHNN can diagnose the efficient inputs, so this feature was applied to determine the efficient L and C elements on TZs, Fc, IL, and RL. Finally, the fabricated filter results are well‐matched with the simulation ones. IL and RL of the suggested filter in operational frequency are −0.09 and −17.26 dB, respectively. The total filter size is only 14.65 mm × 3.5 mm.

Experimental study of the effect of HHO gas injection on pollutants produced by a diesel engine at idle speed

In this study, with the aim of reducing the energy consumption in the production of HHO gas for use in the combustion process of diesel fuel, different modes of gas production were investigated using electrolyzers. According to previous studies, the energy consumption rate of the electrolyzer to produce a high volumetric flow of HHO gas is very high. This high rate will restrict the use of equipment such as high-capacity batteries. The effects of HHO gas injection at the idle speed of the engine at a low temperature were evaluated. Because in this situation, the engine makes high air pollution. The results showed that the percentage of CO, CO2, HC, and NOX gases decreased by 66%, 33%, 38%, and 11%, respectively. On the other hand, the amount of O2 gas in the exhaust increased by 18%. These results were reported for HHO gas injection from 10 to 45 ml/s. The performance of Group Method of Data Handling (GMDH) neural network was desirable in modeling diesel engine pollutants. Because the Root-Mean-Square Error (RMSE) criterion for all evaluated gases is less than 0.32. The GMDH neural network was used for modeling the operation of the diesel engine with HHO supplemental fuel. The GMDH results showed that this artificial network can measure all engine exhaust gases. It can be used as a sensor and virtual simulator for this diesel engine with HHO supplemental fuel.

High‐resolution reference evapotranspiration for arid Egypt: Comparative analysis and evaluation of empirical and artificial intelligence models

AbstractAccurate estimation of evapotranspiration has crucial importance in arid regions like Egypt, which suffers from the scarcity of precipitation and water shortages. This study provides an investigation of the performance of 31 widely used empirical equations and 20 models developed using five artificial intelligence (AI) algorithms to estimate reference evapotranspiration (ET0) to generate gridded high‐resolution daily ET0 estimates over Egypt. The AI algorithms include support vector machine‐radial basis function (SVM‐RBF), random forest (RF), group method of data handling neural network (GMDH‐NN), multivariate adaptive regression splines (MARS), and dynamic evolving neural fuzzy interference system (DENFIS). Daily observations records of 41 stations distributed over Egypt were used to calculate ET0 using FAO56 Penman–Monteith equation as a reference estimate. The multiparameter Kling‐Gupta efficiency (KGE) metric was used as an evaluation metric for its robustness in representing different statistical error/agreement characteristics in a single value. By category, the empirical equations based on radiation performed better in replicating FAO56‐PM followed by temperature‐ and mass‐transfer‐based ones. Ritchie equation was found to be the best overall in Egypt (median KGE 0.76) followed by Caprio (median KGE 0.64), and Penman (median KGE 0.52) equations based on station‐wise ranking. On the other hand, the RF model, having maximum and minimum temperatures, wind speed, and relative humidity as predictors, outperformed other AI algorithms. Overall, the RF model performed the best among all the AI models and empirical equations. The generated 0.10° × 0.10° daily estimates of ET0 enabled the detection of a significant increase of 0.12–0.16 mm·decade−1 in the agricultural‐dependent Nile Delta using the modified Mann–Kendall test and Sen's slope estimator.

A New Decision-Making GMDH Neural Network: Effective for Limited and Fuzzy Data.

This paper presents a new approach to solve multi-objective decision-making (DM) problems based on neural networks (NN). The utility evaluation function is estimated using the proposed group method of data handling (GMDH) NN. A series of training data is obtained based on a limited number of initial solutions to train the NN. The NN parameters are adjusted based on the error propagation training method and unscented Kalman filter (UKF). The designed DM is used in solving the practical problem, showing that the proposed method is very effective and gives favorable results, under limited fuzzy data. Also, the results of the proposed method are compared with some similar methods.

The Influence of Data Length on the Performance of Artificial Intelligence Models in Predicting Air Pollution

Air pollution is one of humanity's most critical environmental issues and is considered contentious in several countries worldwide. As a result, accurate prediction is critical in human health management and government decision-making for environmental management. In this study, three artificial intelligence (AI) approaches, namely group method of data handling neural network (GMDHNN), extreme learning machine (ELM), and gradient boosting regression (GBR) tree, are used to predict the hourly concentration of PM2.5 over a Dorset station located in Canada. The investigation has been performed to quantify the effect of data length on the AI modeling performance. Accordingly, nine different ratios (50/50, 55/45, 60/40, 65/35, 70/30, 75/25, 80/20, 85/15, and 90/10) are employed to split the data into training and testing datasets for assessing the performance of applied models. The results showed that the data division significantly impacted the model's capacity, and the 60/40 ratio was found more suitable for developing predictive models. Furthermore, the results showed that the ELM model provides more precise predictions of PM2.5 concentrations than the other models. Also, a vital feature of the ELM model is its ability to adapt to the potential changes in training and testing data ratio. To summarize, the results reported in this study demonstrated an efficient method for selecting the optimal dataset ratios and the best AI model to predict properly which would be helpful in the design of an accurate model for solving different environmental issues.

A Novel Condition Monitoring Method of Wind Turbines Based on GMDH Neural Network

The safety of power transmission systems in wind turbines is crucial to the wind turbine’s stable operation and has attracted a great deal of attention in condition monitoring of wind farms. Many different intelligent condition monitoring schemes have been developed to detect the occurrence of defects via supervisory control and data acquisition (SCADA) data, which is the most commonly applied condition monitoring system in wind turbines. Normally, artificial neural networks are applied to establish prediction models of the wind turbine condition monitoring. In this paper, an alternative and cost-effective methodology has been proposed, based on the group method of data handling (GMDH) neural network. GMDH is a kind of computer-based mathematical modelling and structural identification algorithm. GMDH neural networks can automatically organize neural network architecture by heuristic self-organization methods and determine structural parameters, such as the number of layers, the number of neurons in hidden layers, and useful input variables. Furthermore, GMDH neural network can avoid over-fitting problems, which is a ubiquitous problem in artificial neural networks. The effectiveness and performance of the proposed method are validated in the case studies.

A new pressure and temperature dependent relative viscosity model for water-Al2O3 nanofluids using GMDH neural network

The introduction of nanoparticles into fluids have been used in various sciences such as; heat transfer, drilling mud, hydraulic fracturing, EOR, and widely in biotechnology science. The dispersion of nanoparticles in the fluid can affect significantly the viscosity of the system. In this study, a self-organizing polynomial neural network based on a group method of data handling (GMDH) is developed to investigate the viscosity of water-based nanofluids regarding temperature and especially pressure changes. The influence of high pressure has not been previously considered in any correlation. The results were estimated using the GMDH model and have higher accuracy than various theoretical models and empirical equations and show the coefficient of determination of 0.999.

Determination of Discharge Distribution in Meandering Compound Channels Using Machine Learning Techniques

Accurate flow rate prediction is essential to analyze flood control, sediment transport, riverbank protection, and so forth. The flow rate distribution becomes even more complicated in compound channels due to the momentum transfer between different subsections across the width of the channel. Conventional channel division methods estimate flow distribution at the main channel and floodplains by assuming a division line with zero apparent shear stress. The article attempts to develop a model to calculate the percentage of discharge in the main channel (%Qmc) using techniques such as Group Method of Data Handling—Neural Network (GMDH-NN) and gene-expression programming (GEP) by incorporating the effects of various geometric and hydraulic parameters. The paper proposes a modified channel division method with a variable-inclined interface, with zero apparent shear force distribution at the channel subsections according to the statistical indices employed to assess these models’ performance in predicting %Qmc. This variable-inclined interface changes its slope according to the channel parameters. The model’s effectiveness is verified by validating with experimental observations by conventional analytical methods.

Enhanced group method of data handling (GMDH) for permeability prediction based on the modified Levenberg Marquardt technique from well log data

Permeability is the key variable for reservoir characterization used for estimating the flow patterns and volume of hydrocarbons. Modern computer advancement has highlighted the use of machine learning approaches such as group method of data handling (GMDH) in predicting permeability. However, the widely employed GMDH has intrinsic problems in its application. Therefore, the objective of this study is to present an enhanced GMDH based modified Levenberg-Marquardt (LM) as an improved alternative to conventional GMDH in predicting permeability from well logs. The study used natural gamma-ray, standard resolution formation density, limited effective porosity, shale volume of rock, and thermal neutron porosity well logs as input variables. Results show that an enhanced method has a reasonable reduction in processing time with high accuracy. Compared to conventional GMDH and backpropagation neural networks (BPNN), the GMDH-LM used 30% less computation time and performed excellently during training with the least error values of 0.092 and 0.018 for RMSE and MAE. Likewise, good results were observed during testing, obtaining the least error values of 0.679 and 0.056 for RMSE and MAE respectively. The modified generalization performance of GMDH-LM makes it an improved form of GMDH and can be adopted as an improved alternative in predicting permeability.

A semantic malware detection model based on the GMDH neural networks

There are several approaches for preventing mobile devices from malware intrusion, but most of them suffer from the insufficient accuracy required for detecting Trojan malware. A combination of semantic and machine learning techniques can be effective in preventing intrusions. In this paper, we have used a hierarchical semantic approach to convert numerical and string data to meaningful values, Subgraph Semantic Homomorphism Coefficient (SSHC) to select optimal features, and Group Method of Data Handling (GMDH) deep neural network (DNN) algorithm to detect malware via a cloud-computing infrastructure. To evaluate our model, Android Trojan Dataset has been used. After evaluation, the accuracy reached 99.91%, which was improved by about 5.25% compared to StormDroid, Drebin, and KuafuDet models. Also, the accuracy was improved by about 10.4% and 31.9% compared to machine learning based approaches of Random Forest (RF), Support Vector Machine (SVM), and K-Nearest Neighbor (KNN), in the state-of-the-art KuafuDet model, respectively.

A hybrid GMDH neural network and logistic regression framework for state parameter–based liquefaction evaluation

The cyclic stress or liquefaction behavior of granular materials is strongly affected by the relative density and confining pressure of the soil. In this study, the state parameter accounting for both relative density and effective stress was used to evaluate soil liquefaction potential. Based on case histories along with the cone penetration test (CPT) database, models for calculating the state parameter using a group method of data handling (GMDH) neural network were developed and recommended according to their performance. The state parameter was then used to develop a state parameter–based probabilistic liquefaction evaluation method using a logistic regression model. From a conservative point of view, the boundary curve of 20% probability of liquefaction was suggested as a deterministic criterion for state parameter–based liquefaction evaluation. Subsequently, a mapping function relating the calculated factor of safety (FS) to the probability of liquefaction (PL) was proposed based on the compiled CPT database. Based on the developed PL–FS function, a new risk criterion associated with the state parameter–based design chart was proposed. Finally, a flowchart of state-based probabilistic liquefaction evaluation and quality control for ground-improvement projects was presented for the benefit of practitioners.