Application of machine learning algorithms for forecasting labour demand in the metallurgical industry of the east Kazakhstan region

The use of recycled glass in the concrete mix instead of natural coarse aggregates and supplemental cementitious material has several advantages, including the conservation of natural resources, the reduction of CO2 emissions, and cost savings. However, due to their qualities, the mechanical properties of concrete containing Ground Glass Particles (GGP) differ from those of natural aggregates concrete. As a result, assessing the compressive strength (CS) of concrete with GGP is crucial. Therefore, this paper proposes the hybrid Machine Learning (ML) model including the Gradient Boosting (GB) and Bayesian optimization (BO) algorithms for predicting the compressive strength of concrete containing GGP. The hybrid ML model is developed and validated based on the training dataset (70% of the data) and the test dataset (30% of the remaining data), respectively. The performance of hybrid ML model is evaluated by three criteria, such as the Pearson correlation coefficient (R), Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The K-Fold Cross-Validation technique is also used to verify the reliability of the hybrid ML model). The best performance of the hybrid ML model is determined with the R = 0.9843, RMSE = 1.7256 (MPa), and MAE = 1.3154 (MPa) for training dataset and R = 0.9784, RMSE = 2.4338 (MPa) and MAE = 1.9618 (MPa) for testing dataset. Based on the best hybrid ML model, the sensitivity analysis including SHapley Additive exPlanation (SHAP) and Partial Dependence Plots (PDP) 2D are investigated to obtain an in-depth examination of each individual input variable on the predicted compressive strength of concrete contaning GGP. The sensitivity analysis shows that four factors, such as curing age, surface area, TiO2, and temperature have the most effect on the compressive strength of concrete containing GGP.

Axial strength prediction of steel tube confined concrete columns using a hybrid machine learning model

River water temperature prediction using hybrid machine learning coupled signal decomposition: EWT versus MODWT

Ultra-high-performance concrete (UHPC) is a sustainable construction material; it can be applied as a substitute for cement concrete. Artificial intelligence methods have been used to evaluate concrete composites to reduce time and money in the construction industries. So, this study applied machine learning (ML) and hybrid ML approaches to predict the compressive and flexural strength of UHPC. A dataset of 626 compressive strength and 317 flexural strength data points was collected from the published research articles, where fourteen important variables were selected as input parameters for the analysis of hybrid ML and ML algorithms. This research used XGBoost, LightGBM, and hybrid XGBoost- LightGBM algorithms to predict UHPC materials. Grid search (GS) techniques were used to adjust model hyper-parameters in search of improved high accuracy and efficiency. ML and hybrid ML models were train, and the test stage utilized statistical assessments such as coefficient of determination (R-square), root mean square error (RMSE), mean absolute error (MAE), and coefficient of efficiency (CE). The results presented hybrid ML algorithm was superior to the XGBoost and LightGBM algorithms in terms of R-square and RMSE values for both compressive and flexural strength prediction. A hybrid ML model and two ML models showed CS considerable R-square values above 0.94 at the testing stages and just over 0.97 at the training phase. Hybrid ML model performance accuracy for CS prediction R-square value found that almost 0.996 for training and 0.963 for testing phases. At the same time, the FS prediction result showed that the R-square value of the Hybrid ML model and two traditional ML models were found at almost 0.95 for the training phase and around 0.81 for the testing phase. But among them, the hybrid XGB-LGB model prediction performance was high accuracy and lowest error for CS and FS of UHPC trained and its hyperparameters optimized. Additionally, the SHAP investigation reveals the impact and relationship of the input variables with the output variables. SHAP analysis outcome reveals that curing age and steel fiber content input parameter had the highest positive impact on compressive strength and flexural strength of UHPC.

Investigating vocational educators' knowledge-based teaching skills across China's Vocational Education (VE) institutions, this research focuses on the practical use of Machine Learning (ML) algorithms. Instructors' efficacy must be evaluated, and this work addresses the gap. VE performs an essential role in connecting learning abilities with the demands of industry. The investigation plans on developing an adaptable, subjective assessment technique that extends within the boundaries of conventional subjective evaluation methods using modern ML techniques such as Support Vector Machines (SVM), Decision Trees (DT), and Neural Networks (NN). Each ML model's accuracy, reliability, and feasibility have been determined using data collected from 120 vocational educators encompassing various fields and regions. Researchers predict that our findings will provide perspective on how to improve vocational education settings' teaching methods and governance.

This study discusses the application of machine learning algorithms for the prediction of Microsoft stock prices using historical data for five years. Preprocess of the data was done by treating missing values, creating lag features, and normalizing the data for better model performance. To the price of close and momentum, various models were trained, including Linear Regression, Decision Tree, Random Forest, Support Vector Regression, and Gradient Boosting Regressor. Based on the experimental results, Linear Regression achieved the best performance in closing price prediction, recording a Coefficient of Determination (R²) of 0.91, Mean Squared Error (MSE) of 45.77, Mean Absolute Error (MAE) of 4.64, and Mean Absolute Percentage Error (MAPE) of 0.01. For momentum prediction, Linear Regression again outperformed other models, achieving R² = 0.51, MSE = 21.25, MAE = 2.79, and MAPE = 2.05. And other models showed much weaker explanatory power. When predicting the Relative Strength Index (RSI) classification, Gradient Boosting delivered the best overall performance, achieving Accuracy = 1.00, F1 score = 1.00, Cross-Validation (CV) mean accuracy = 0.998, and CV standard deviation (CV std) = 0.02. Although other models such as Linear Regression, Logistic Regression, Support Vector Classifier, and Random Forest achieved strong results, none matched the superior performance of Gradient Boosting.

The rapid advancements in computing technologies, especially quantum computing, pose significant challenges to traditional encryption methods, compelling the need for more robust, adaptive, and scalable solutions. Hybrid machine learning (ML) models have emerged as a promising approach to address these challenges, offering enhanced security, performance, and scalability. This book chapter explores the intersection of hybrid ML models and encryption methodologies, focusing on how these models can transform data encryption techniques for secure communication. By integrating various ML techniques such as supervised, unsupervised, and reinforcement learning, hybrid models provide adaptive encryption strategies that can dynamically respond to emerging threats and evolving system requirements. The chapter delves into the application of hybrid ML models in quantum-safe encryption, key management systems, and real-time adaptive encryption, showcasing case studies that demonstrate their practical impact in securing data in both traditional and quantum computing environments. Through comprehensive analysis, this chapter highlights the potential of hybrid ML models to optimize encryption efficiency, enhance key exchange protocols, and ensure the scalability of encryption systems, paving the way for a secure and future-proof communication infrastructure.

Share prices are a critical factor in a stock index’s worth but are never constant. Thus, an effective method of predicting share prices is needed. This is where machine learning comes in. This research discusses the applicability of machine learning algorithms, precisely long short-term memory, artificial neural networks, and linear regression in predicting share prices. Additionally, this research goes in-depth, explaining how each algorithm functions. These three algorithms were implemented using the financial dataset of the S&P 500, one of the more known stock indices out there. Data was collected from Yahoo Finance for 34 years, from 1990 to 2023. Then, the algorithms mentioned were used to train a model using the collected dataset. All three algorithms were measured using three performance metrics: mean absolute error, R-squared score, and mean absolute percentage error. The final implementation involved training them by only using 1-day lagged features to create a model that can predict the next day's closing price. All the algorithms performed considerably well, with linear regression being the best, followed by artificial neural networks and long short-term memory being the worst. Finally, the implemented algorithms were used to predict the closing prices of other stock indices, NASDAQ and Hang Seng Index. All algorithms performed well and followed the same trend, wherein linear regression performed the best and long- and short-term memory the worst. Future research should be conducted to explore the possibilities of utilizing lagged features along with external features like GDP growth rate, political trends, etc.

Prediction of geochemical concentration values is essential in mineral exploration as it plays a principal role in the economic section. In this paper, four regression machine learning (ML) algorithms, such as K neighbor regressor (KNN), support vector regressor (SVR), gradient boosting regressor (GBR), and random forest regressor (RFR), have been trained to build our proposed hybrid ML (HML) model. Three metric measurements, including the correlation coefficient, mean absolute error (MAE), and means squared error (MSE), have been selected for model prediction performance. The final prediction of Pb and Zn grades is achieved using the HML model as they outperformed other algorithms by inheriting the advantages of individual regression models. Although the introduced regression algorithms can solve problems as single, non-complex, and robust regression models, the hybrid techniques can be used for the ore grade estimation with better performance. The required data are gathered from in situ soil. The objective of the recent study is to use the ML model’s prediction to classify Pb and Zn anomalies by concentration-area fractal modeling in the study area. Based on this fractal model results, there are five geochemical populations for both cases. These elements’ main anomalous regions were correlated with mining activities and core drilling data. The results indicate that our method is promising for predicting the ore elemental distribution.

Developing precise soft computing methods for groundwater management, which includes quality and quantity, is crucial for improving water resources planning and management. In the past 20 years, significant progress has been made in groundwater management using hybrid machine learning (ML) models as artificial intelligence (AI). Although various review articles have reported advances in this field, existing literature must cover groundwater management using hybrid ML. This review article aims to understand the current state-of-the-art hybrid ML models used for groundwater management and the achievements made in this domain. It includes the most cited hybrid ML models employed for groundwater management from 2009 to 2022. It summarises the reviewed papers, highlighting their strengths and weaknesses, the performance criteria employed, and the most highly cited models identified. It is worth noting that the accuracy was significantly enhanced, resulting in a substantial improvement and demonstrating a robust outcome. Additionally, this article outlines recommendations for future research directions to enhance the accuracy of groundwater management, including prediction models and enhance related knowledge.

This study investigates the dry reformation of methane (DRM) over Ni/Al2O3 catalysts in a dielectric barrier discharge (DBD) non-thermal plasma reactor. A novel hybrid machine learning (ML) model is developed to optimize the plasma-catalytic DRM reaction with limited experimental data. To address the non-linear and complex nature of the plasma-catalytic DRM process, the hybrid ML model integrates three well-established algorithms: regression trees, support vector regression, and artificial neural networks. A genetic algorithm (GA) is then used to optimize the hyperparameters of each algorithm within the hybrid ML model. The ML model achieved excellent agreement with the experimental data, demonstrating its efficacy in accurately predicting and optimizing the DRM process. The model was subsequently used to investigate the impact of various operating parameters on the plasma-catalytic DRM performance. We found that the optimal discharge power (20 W), CO2/CH4 molar ratio (1.5), and Ni loading (7.8 wt%) resulted in the maximum energy yield at a total flow rate of ∼51 mL/min. Furthermore, we investigated the relative significance of each operating parameter on the performance of the plasma-catalytic DRM process. The results show that the total flow rate had the greatest influence on the conversion, with a significance exceeding 35% for each output, while the Ni loading had the least impact on the overall reaction performance. This hybrid model demonstrates a remarkable ability to extract valuable insights from limited datasets, enabling the development and optimization of more efficient and selective plasma-catalytic chemical processes.

Biomass is an excellent source of green energy with numerous benefits such as abundant availability, net carbon zero, and renewable nature. However, the conventional methods of biomass combustion are polluting and poor efficiency processes. Biomass gasification overcomes these challenges and provides a sustainable method for the supply of greener fuel in the form of producer gas. The producer gas can be employed as a gaseous fuel in compression ignition engines in dual‐fuel systems. The biomass gasification process is a complex as well as a nonlinear process that is highly dependent on the ambient environment, type of biomass, and biomass composition as well as the gasification medium. This makes the modeling of such systems quite difficult and time‐consuming. Modern machine learning (ML) techniques offer the use of experimental data as a convenient approach to modeling and forecasting such systems. In the present study, two modern and highly efficient ML techniques, random forest (RF) and AdaBoost, were employed for this purpose. The outcomes were employed with results of a baseline method, i.e., linear regression. The RF could forecast the hydrogen yield with R2 as 0.978 during model training and 0.998 during the model test phase. AdaBoost ML was close behind with R2 at 0.948 during model training and 0.842 during the model test phase. The mean squared error was as low as 0.17 and 0.181 during model training and testing, respectively. In the case of the low heating value model, during model testing, the R2 was 0.971 and RF and AdaBoost, respectively, during model training and 0.842 during the model test phase. Both ML techniques provided excellent results compared to linear regression, but RFt was the best among all three.

The principal purpose of this work is to develop three hybrid machine learning (ML) algorithms, namely ANFIS-RCSA, ANFIS-CA, and ANFIS-SFLA which are a combination of adaptive neuro-fuzzy inference system (ANFIS) with metaheuristic optimization techniques such as real-coded simulated annealing (RCSA), cultural algorithm (CA) and shuffled frog leaping algorithm (SFLA), respectively, to predict the critical buckling load of I-shaped cellular steel beams with circular openings. For this purpose, the existing database of buckling tests on I-shaped steel beams were extracted from the available literature and used to generate the datasets for modeling. Eight inputs, considered as independent variables, including the beam length, beam end-opening distance, opening diameter, inter-opening distance, section height, web thickness, flange width, and flange thickness, as well as one output of the critical buckling load of cellular steel beams considered as a dependent variable, were used in the datasets. Three quality assessment criteria, namely correlation coefficient (R), root mean squared error (RMSE) and mean absolute error (MAE) were employed for assessment of three developed hybrid ML models. The obtained results indicate that all three hybrid ML models have a strong ability to predict the buckling load of steel beams with circular openings, but ANFIS-SFLA (R = 0.960, RMSE = 0.040 and MAE = 0.017) exhibits the best effectiveness as compared with other hybrid models. In addition, sensitivity analysis was investigated and compared with linear statistical correlation between inputs and output to validate the importance of input variables in the models. The sensitivity results show that the most influenced variable affecting beam buckling capacity is the beam length, following by the flange width, the flange thickness, and the web thickness, respectively. This study shows that the hybrid ML techniques could help in establishing a robust numerical tool for beam buckling analysis. The proposed methodology is also promising to predict other types of failure, as well as other types of perforated beams.

This paper examines how linear regression in machine learning enables the prediction of air-ground path loss through environmental parameters such as temperature, humidity, and atmospheric pressure measurements. The paper demonstrates that temperature plays the most significant role in determining path loss, while humidity and atmospheric pressure contribute at a lower level. A high level of accuracy defines the linear regression model, which demonstrated efficient path loss prediction through a Mean Absolute Error (MAE) of 0.2995. The model demonstrates effective capabilities for system improvements during changing atmospheric conditions because the trend line shows the smooth progression of predicted and actual values. A hybrid model produced enhanced prediction accuracy when particle swarm optimization and gradient-boosting regressor parameters were optimized to establish the new model system. The optimized model substantially declined MAE to 0.0435, which verified its improved predictive capacity regarding absolute path loss values. A performance-maximized model resulted from tuning relevant parameters to set n_estimators equal to 56, learning rate to 0.1, and max_depth to 9. The optimized model accurately predicts path loss in communication networks, preparing it for on-site deployment. This research serves as a basis for further investigation, integrating other environmental elements, including wind speed, rainfall and elevation levels, and testing alternative state-of-the-art machine learning methods. Future improvements in these procedures can boost the flexibility and reliability of networks with an emphasis on air-ground systems. Research findings indicate that PSO-GBR hybrid models possess a high potential for path loss prediction, creating new possibilities for future air-ground communication systems and emerging technologies such as low-altitude satellites, air taxis, and unmanned aerial vehicles (UAVs).

The object of this research is the impact of technological changes on the dynamics of economic development of enterprises. The main hypothesis is the assumption of the presence of such an impact for a significant number of companies. The implementation of this research made it possible to make a certain contribution to the process of solving the problem of finding ways to accelerate the economic development of business entities. At the same time, technological changes were divided into three groups, namely: resource-saving; changes that ensure the improvement of the quality of the enterprise's products; changes that ensure the improvement of management, sales and other processes at enterprises. A methodological approach to assessing the impact of technological changes on the dynamics of economic development of companies was also developed. This approach involves the implementation of two main methods of assessment, namely: establishing the presence or absence of such an impact and determining the magnitude of the impact of technological changes on the dynamics of economic development of enterprises. The testing of the developed tool on a sample of industrial enterprises showed that the impact of technological changes on the dynamics of their economic development exists and is statistically significant. At the same time, the average impact of technological changes on the growth of financial results of enterprises is quite high. In particular, the average values of the indicator of the impact of technological changes on the net profit of those enterprises that have undergone at least two types of such changes, by type of economic activity, range from 11.25% to 13.32%. Since a significant number of the enterprises studied have not carried out technological changes in recent years, at least some of these enterprises may have significant potential to accelerate their economic development. The developed toolkit for assessing the impact of technological changes on the dynamics of economic development of enterprises can be used to establish the presence and extent of such an impact both at the level of an individual company and at the industry level. This will allow owners and managers of enterprises to increase the validity of the strategies for technological renewal of these enterprises.