Search

Travel mode choice (TMC) prediction is crucial for transportation planning. Most previous studies have focused on TMC in adults, whereas predicting TMC in children has received less attention. On the other hand, previous children's TMC prediction studies have generally focused on home-to-school TMC. Hence, LIGHT GRADIENT BOOSTING MACHINE (LGBM), as a robust machine learning method, is applied to predict children's TMC and detect its determinants since it can present the relative influence of variables on children's TMC. Nonetheless, the use of machine learning introduces its own challenges. First, these methods and their performance are highly dependent on the choice of "hyperparameters". To solve this issue, a novel technique, called multi-objective hyperparameter tuning (MOHPT), is proposed to select hyperparameters using a multi-objective metaheuristic optimization framework. The performance of the proposed technique is compared with conventional hyperparameters tuning methods, including random search, grid search, and "Hyperopt". Second, machine learning methods are black-box tools and hard to interpret. To overcome this deficiency, the most influential parameters on children's TMC are determined by LGBM, and logistic regression is employed to investigate how these parameters influence children's TMC. The results suggest that MOHPT outperforms conventional methods in tuning hyperparameters on the basis of prediction accuracy and computational cost. Trip distance, "walkability" and "bikeability" of the origin location, age, and household income are principal determinants of child mode choice. Furthermore, older children, those who live in walkable and bikeable areas, those belonging low-income groups, and short-distance travelers are more likely to travel by sustainable transportation modes.

The limited nature of fossil resources and their unsustainable characteristics have led to increased interest in renewable sources. However, significant work remains to be carried out to fully integrate these systems into existing power distribution networks, both technically and legally. While reliability holds great potential for improving energy production sustainability, the dependence of solar energy production plants on weather conditions can complicate the realization of consistent production without incurring high storage costs. Therefore, the accurate prediction of solar power production is vital for efficient grid management and energy trading. Machine learning models have emerged as a prospective solution, as they are able to handle immense datasets and model complex patterns within the data. This work explores the use of metaheuristic optimization techniques for optimizing recurrent forecasting models to predict power production from solar substations. Additionally, a modified metaheuristic optimizer is introduced to meet the demanding requirements of optimization. Simulations, along with a rigid comparative analysis with other contemporary metaheuristics, are also conducted on a real-world dataset, with the best models achieving a mean squared error (MSE) of just 0.000935 volts and 0.007011 volts on the two datasets, suggesting viability for real-world usage. The best-performing models are further examined for their applicability in embedded tiny machine learning (TinyML) applications. The discussion provided in this manuscript also includes the legal framework for renewable energy forecasting, its integration, and the policy implications of establishing a decentralized and cost-effective forecasting system. [ABSTRACT FROM AUTHOR]

Rockburst is a major challenge to hard rock engineering at great depth. Accurate and timely assessment of rockburst risk can avoid unnecessary casualties and property losses. Despite the existence of various methods for rockburst assessment, there remains an urgent need for a comprehensive and reliable criterion that is easy to both apply and interpret. Developing a new rockburst criterion based on simple parameters can potentially fill this gap. With its advantages, this criterion can facilitate a more effective and efficient prediction of rockburst potential, thereby contributing significantly to enhancing safety measures. In this paper, combined with the internal and external factors of rockburst, four control variables (i.e., integrity index, stress index, brittleness index, and elastic energy index) were selected to be incorporated into a comprehensive rockburstability index (RBSI). Based on 116 sets of rockburst cases, the rockburst potential was accurately quantified and predicted using the categorical boosting (CatBoost) model and the nature-inspired metaheuristic African vultures optimization algorithm (AVOA). In its performance validation, the criterion achieved the highest accuracy of 90.48%, verifying the reliability and effectiveness of the proposed RBSI criterion. Additionally, an interpretive method was applied to analyze the variable influence on the criterion, facilitating the explanation of predictions and the analysis of the formula's robustness under different conditions. In general, compared with existing criterion methods involving relevant indicators, the newly proposed RBSI criterion enhances the accuracy of rockburst potential prediction, and it can effectively and swiftly evaluate the preliminary risk of rockburst. Lastly, a graphical user interface was developed to provide a clear visualization of the assessment of rockburst potential. [ABSTRACT FROM AUTHOR]

Modern civilization is heavily dependent on energy, which burdens the energy sector. Therefore, a highly accurate energy consumption forecast is essential to provide valuable information for efficient energy distribution and storage. This study proposed a hybrid deep learning model, called I-CNN-JS, by incorporating a jellyfish search (JS) algorithm into an ImageNetwinning convolutional neural network (I-CNN) to predict weekahead energy consumption. First, numerical data were encoded into grayscale images for input of the proposed model, showcasing the novelty of using image data for analysis. Second, a newly developed metaheuristic optimization algorithm, JS, was used to improving model accuracy. Results showed that the proposed method outperformed conventional numerical input methods. The optimized model yielded a mean absolute percentage error improvement of 0.5% compared to the default models, indicating that JS is a promising method for achieving the optimal hyperparameters. Sensitivity analysis further evaluated the impact of image pixel orientation on performance model. [ABSTRACT FROM AUTHOR]

Chloride corrosion significantly impacts the durability of reinforced concrete (RC) structures. Traditional evaluation methods are time-consuming and expensive. Machine Learning (ML) offers a promising alternative, providing efficient and accurate predictions. This review explores recent ML advancements in assessing corrosion in RC structures. Various algorithms, such as Artificial Neural Networks (ANNs), Gene Expression Programming (GEP), Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM) and Ensemble Learning, have shown potential in estimating corrosion processes, predicting material properties, and evaluating structural durability. Future research should focus on integrating ML with physical models to enhance robustness and reliability in service life prediction. This review summarizes current trends, challenges, and the future potential of ML in predicting chloride ingress and its impact on concrete durability. [ABSTRACT FROM AUTHOR]

This study explores the realm of time series forecasting, focusing on the utilization of Recurrent Neural Networks (RNN) to detect abnormal cardiovascular rhythms in Electrocardiogram (ECG) signals. The principal objective is to optimize RNN performance by finely tuning hyperparameters, a complex task with known NP-hard complexity. To address this challenge, the study employs metaheuristic algorithms, specialized problem-solving techniques crafted for navigating intricate and non-deterministic optimization landscapes. Additionally, a refined algorithm is introduced to overcome limitations inherent in the original approach. This modified algorithm exhibits significant improvements, surpassing its predecessor in identifying anomalous cardiovascular rhythms within ECG signals. The most successful optimized model achieves an accuracy of 99.26%, outperforming models optimized by other contemporary metaheuristics assessed in the study. Further experimentation extends the initial inquiry by exploring the capabilities of Long Short-Term Memory (LSTM) models augmented by attention layers. In this extension, the best models demonstrate an accuracy of 99.83%, surpassing the original RNN models. These findings underscore the crucial importance of refining machine learning models and emphasize the potential for substantial advancements in healthcare through innovative algorithmic approaches. [ABSTRACT FROM AUTHOR]

Shear strength parameters play a crucial role in the design and construction of rock slopes, underground openings, tunnels, excavations, and foundations. Determination of cohesion (c) and angle of internal friction (φ) through laboratory tests (e.g., triaxial tests) and in situ tests is time-consuming and expensive; therefore, it is valuable to accurately predict the shear strength parameters of rock materials using less expensive and more reliable methods. In this study, 4 artificial intelligence models, namely group method of data handling (GMDH), gene expression programming (GEP), bidirectional long short-term memory network (BILSTM), and random forest (RF), are proposed to predict the rock shear strength parameters c and φ. A database of 199 sets of experimental data from 4 rock types was used to construct the proposed models. The results show that the accuracy of the RF model is superior to the other three models, with coefficients of determination of 0.9933 (c) and 0.9727 (φ) on all datasets, respectively. To further improve the prediction accuracy of the RF model, six metaheuristic algorithms, namely particle swarm optimization (PSO), bald eagle search (BES), marine predators algorithm (MPA), northern goshawk optimization (NGO), golden jackal optimization (GJO), and dung beetle optimizer (DBO), were used to optimize the hyperparameters of the model. The results show that the accuracy of the six hybrid RF models is higher than that of the single RF model. Among the 6 hybrid RF models, the hybrid DBO–RF model is superior to the other hybrid RF models. In addition, the factors affecting the shear strength of rock materials are analyzed through the parameter sensitivity analysis. Highlights: Four artificial intelligence models are used to predict the shear strength parameters of rock materials. Six metaheuristic algorithms are used to optimize the hyperparameters of the random forest model. The prediction performance of the six hybrid random forest models is compared. The factors influencing the shear strength of rock materials are analyzed by parameter sensitivity analysis. [ABSTRACT FROM AUTHOR]

AbstractThe present research examines the utilization of metaheuristic optimization algorithms to enhance the predictive accuracy of the Extreme Gradient Boosting (XGBoost) machine learning model in predicting the capacity for punching shear of Fiber Reinforced Polymer (FRP) reinforced concrete slabs in the absence of shear reinforcement. A database including 610 experimental results is constructed, and the efficiency of twenty metaheuristic algorithms is assessed in optimizing the hyperparameters of the XGBoost model. The enhanced models undergo evaluation and comparison with existing prediction algorithms, showcasing their better accuracy. The study reveals that the DE, MVO, EOA, and VCS algorithms yield the highest precision among the investigated optimization techniques out of which VCS model performed better. Furthermore, the significance of input parameters on the punching shear capacity is analyzed by implementing the Shapley Additive Explanation (SHAP) method. The findings emphasize the potential of metaheuristic optimization in enhancing the estimating abilities of machine learning models for complex engineering problems and provide valuable insights into the key factors influencing the punching shear strength of concrete slabs reinforced with FRP. [ABSTRACT FROM AUTHOR]

Accurately predicting wind speeds is of great significance in various engineering applications, such as the operation of high-speed trains. Machine learning models are effective in this field. However, existing studies generally provide deterministic predictions and utilize decomposition techniques in advance to enhance predictive performance, which may encounter data leakage and fail to capture the stochastic nature of wind data. This work proposes an advanced framework for the prediction and early warning of wind speeds by combining the optimized gated recurrent unit (GRU) and adaptive kernel density estimator (AKDE). Firstly, 12 samples (26,280 points each) were collected from an extensive open database. Three representative metaheuristic algorithms were then employed to optimize the parameters of diverse models, including extreme learning machines, a transformer model, and recurrent networks. The results yielded an optimal selection using the GRU and the crested porcupine optimizer. Afterwards, by using the AKDE, the joint probability density and cumulative distribution function of wind predictions and related predicting errors could be obtained. It was then applicable to calculate the conditional probability that actual wind speed exceeds the critical value, thereby providing probabilistic-based predictions in a multilevel manner. A comparison of the predictive performance of various methods and accuracy of subsequent decisions validated the proposed framework. [ABSTRACT FROM AUTHOR]

Accurate rock mass quality classification is crucial for the design and construction of underground projects. Traditional methods often rely on expert experience, introducing subjectivity, and struggle with complex geological conditions. Machine learning algorithms have improved this issue, but obtaining complete rock mass quality datasets is often difficult due to high cost and complex procedures. This study proposed a hybrid XGBoost model for predicting rock mass quality using incomplete datasets. The zebra optimization algorithm (ZOA) and Bayesian optimization (BO) were used to optimize the hyperparameters of the model. Data from various regions and types of underground engineering projects were utilized. Adaptive synthetic (ADASYN) oversampling addressed class imbalance. The model was evaluated using metrics including accuracy, Kappa, precision, recall, and F1-score. The ZOA-XGBoost model achieved an accuracy of 0.923 on the test set, demonstrating the best overall performance. Feature importance analysis and individual conditional expectation (ICE) plots highlighted the roles of RQD and UCS in predicting rock mass quality. The model's robustness with incomplete data was verified by comparing its performance with other machine learning models on a dataset with missing values. The ZOA-XGBoost model outperformed other models, proving its reliability and effectiveness. This study provides an efficient and objective method for rock mass quality classification, offering significant value for engineering applications. [ABSTRACT FROM AUTHOR]

Feature selection (FS) is a significant dimensionality reduction technique in machine learning and data mining that is adept at managing high-dimensional data efficiently and enhancing model performance. Metaheuristic algorithms have become one of the most promising solutions in FS owing to their powerful search capabilities as well as their performance. In this paper, the novel improved binary walrus optimizer (WO) algorithm utilizing the golden sine strategy, elite opposition-based learning (EOBL), and population regeneration mechanism (BGEPWO) is proposed for FS. First, the population is initialized using an iterative chaotic map with infinite collapses (ICMIC) chaotic map to improve the diversity. Second, a safe signal is obtained by introducing an adaptive operator to enhance the stability of the WO and optimize the trade-off between exploration and exploitation of the algorithm. Third, BGEPWO innovatively designs a population regeneration mechanism to continuously eliminate hopeless individuals and generate new promising ones, which keeps the population moving toward the optimal solution and accelerates the convergence process. Fourth, EOBL is used to guide the escape behavior of the walrus to expand the search range. Finally, the golden sine strategy is utilized for perturbing the population in the late iteration to improve the algorithm's capacity to evade local optima. The BGEPWO algorithm underwent evaluation on 21 datasets of different sizes and was compared with the BWO algorithm and 10 other representative optimization algorithms. The experimental results demonstrate that BGEPWO outperforms these competing algorithms in terms of fitness value, number of selected features, and F1-score in most datasets. The proposed algorithm achieves higher accuracy, better feature reduction ability, and stronger convergence by increasing population diversity, continuously balancing exploration and exploitation processes and effectively escaping local optimal traps. [ABSTRACT FROM AUTHOR]

Since cooling load estimation directly impacts air conditioning control and chiller optimization, it is essential for increasing the energy efficiency of cooling systems. Machine learning outshines traditional regression analysis by efficiently managing vast datasets and discerning complex patterns influenced by various factors like occupancy, building materials, and meteorology. These capabilities greatly enhance building management and energy optimization. The primary objective of this study is to introduce a framework based on ML algorithms to accurately predict cooling loads in buildings. The Decision Tree model was chosen as the core algorithm for this purpose. Furthermore, as an innovative approach, four metaheuristic algorithms – namely, the Improved Arithmetic Optimization Algorithm, Prairie Dog Optimization, Covariance Matrix Adaptation Evolution Strategy, and Coyote Optimization Algorithm – were employed to enhance the predictive capabilities of the Decision Tree model. The dataset which utilized in this study derived from previous studies, the data comprised of eight input parameters, including Relative Compactness, Surface Area, Wall Area, Roof Area, Overall Height, Orientation, Glazing Area, and Glazing Area Distribution. With an astonishing R2 value of 0.995 and a lowest Root Mean Square Error value of 0.660, the DTPD (DT + PDO) model performs exceptionally well. These astounding findings demonstrate the DTPD model's unmatched precision in forecasting the results of cooling loads and point to its potential for useful implementation in actual building management situations. Properly predicting and managing cooling loads ensures that indoor environments remain comfortable and healthy for occupants. Maintaining optimal temperature and humidity levels not only enhances comfort but also supports good indoor air quality. [ABSTRACT FROM AUTHOR]

Deep-seated landslides, becoming increasingly frequent due to changing climate patterns, pose significant risks to human life and infrastructure. This research contributes to developing predictive early warning systems for deep-seated slope displacements, employing advanced computational models for environmental risk management. Our novel framework integrates machine learning, time series deep learning, and convolutional neural networks (CNN), enhanced by the Age of Exploration-Inspired Optimizer (AEIO) algorithm. Our approach demonstrates exceptional forecasting capabilities by utilizing eight years of comprehensive data—including displacement, groundwater levels, and meteorological information from the Lushan Mountain region in Taiwan. The AEIO-MobileNet model stands out for its precision in predicting imminent slope displacements with a mean absolute percentage error (MAPE) of 2.81 %. These advancements significantly enhance geohazard informatics by providing reliable and efficient landslide risk assessment and management tools. These safeguard road networks, construction projects, and infrastructure within vulnerable slope areas. [ABSTRACT FROM AUTHOR]

Transmission of information is an essential component in an IoT device for sending, receiving, and collecting data. The Smart devices in IoT architecture are designed as physical devices linked with computing resources that can connect and communicate with another smart device through any medium and protocol. Communication among various smart devices is a challenging task to exchange information and to guarantee the information reaches the destination entirely in real-time in the same order as sent without any data loss. Thus, this article proposes the novel Bat-based Deep Belief Neural framework (BDBN) method for the air pollution monitoring scheme. The reliability of the proposed system has been tested under the error condition in the transport layer and is validated with the conventional methods in terms of Accuracy, Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Pearson correlation coefficient (r), Coefficient of determination (R2) and Error rate.

A unique hybrid meta-heuristic combining the salp swarm algorithm and the sine cosine algorithm (SSCA) is established in this study to improve convergence speed while outperforming existing conventional algorithms. The sine cosine position equations are utilized to update the position of the salp leader in search space while a weighting factor updates the position of the salp follower so that the best and possible optimal solutions are obtained using the sine or cosine and weighting function. Particle swarm optimization PSO inspires this weighting factor. Each salp uses the information-sharing approach of sine and cosine functions during this process to strengthen their exploration and exploitation abilities. The goal of incorporating modifications to the salp swarm optimizer algorithm is to help the standard approach avoid premature convergence that leads the search to the most likely search space. The proposed algorithm is tested on classical optimization benchmark functions and eight real engineering applications. The goal is to investigate and validate the SSCA's proper behaviour while finding the optimum solutions. The results of the comparison demonstrated that the SSCA method achieves the best accuracies. [ABSTRACT FROM AUTHOR]

Accurate rock mass quality classification is crucial for the design and construction of underground projects. Traditional methods often rely on expert experience, introducing subjectivity, and struggle with complex geological conditions. Machine learning algorithms have improved this issue, but obtaining complete rock mass quality datasets is often difficult due to high cost and complex procedures. This study proposed a hybrid XGBoost model for predicting rock mass quality using incomplete datasets. The zebra optimization algorithm (ZOA) and Bayesian optimization (BO) were used to optimize the hyperparameters of the model. Data from various regions and types of underground engineering projects were utilized. Adaptive synthetic (ADASYN) oversampling addressed class imbalance. The model was evaluated using metrics including accuracy, Kappa, precision, recall, and F1-score. The ZOA-XGBoost model achieved an accuracy of 0.923 on the test set, demonstrating the best overall performance. Feature importance analysis and individual conditional expectation (ICE) plots highlighted the roles of RQD and UCS in predicting rock mass quality. The model’s robustness with incomplete data was verified by comparing its performance with other machine learning models on a dataset with missing values. The ZOA-XGBoost model outperformed other models, proving its reliability and effectiveness. This study provides an efficient and objective method for rock mass quality classification, offering significant value for engineering applications.

Although seismic response predictions are widely used for engineering structures, their applications in electrical equipment are rare. Overstressing at the bottom of the porcelain insulators during seismic events has made power transformer bushings in substations prone to failure. Thus, this paper proposed and compared six integrated machine learning (ML) models for seismic stress response predictions for porcelain transformer bushings using easily monitored acceleration responses. Metaheuristic algorithms such as particle swarm optimization were employed for architecture tuning. Prediction accuracies for stress response values and classifications were evaluated. Finally, shaking table tests and simulation analyses for a 1100 kV bushing were implemented to validate the accuracy of the six ML models. The results indicated that the proposed ML models can quickly forecast the maximum stress experienced by a porcelain bushing during earthquakes. Swarm intelligence evolutionary technologies could quickly and automatically aid in the retrofitting of architecture for the ML models. The K-nearest neighbor regression model had the best level of prediction accuracy among the six selected ML models for experimental and simulation validations. ML prediction models have clear benefits over frequently used seismic analytical techniques in terms of speed and accuracy for post-earthquake emergency relief in substations.

Heart disease is a primary cause of death worldwide and is notoriously difficult to cure without a proper diagnosis. Hence, machine learning (ML) can reduce and better understand symptoms associated with heart disease. This study aims to develop a framework for the automatic and accurate classification of heart disease utilizing machine learning algorithms, grid search (GS), and the Aquila optimization algorithm. In the proposed approach, feature selection is used to identify characteristics of heart disease by using a method for dimensionality reduction. First, feature selection is accomplished with the help of the Aquila algorithm. Then, the optimal combination of the hyperparameters is selected using grid search. The experiments were conducted with three datasets fromKaggle: TheHeart Failure Prediction Dataset, Heart Disease Binary Classification, and Heart Disease Dataset. Two classes can be distinguished: diseased and healthy (i.e., uninfected). The Histogram Gradient Boosting (HGB) classifier produced the highest Weighted Sum Metric (WSM) scores of 98.65% concerning the Heart Failure Prediction Dataset. In contrast, the Decision Tree (DT) machine learning classifier had the highestWSMscores of 87.64% concerning theHeart Disease Health Indicators Dataset. Measures of accuracy, specificity, sensitivity, and other metrics are used to evaluate the proposed approach. The presented method demonstrates superior performance compared to different state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

After the excavation of the roadway, the original stress balance is destroyed, resulting in the redistribution of stress and the formation of an excavation damaged zone (EDZ) around the roadway. The thickness of EDZ is the key basis for roadway stability discrimination and support structure design, and it is of great engineering significance to accurately predict the thickness of EDZ. Considering the advantages of machine learning (ML) in dealing with high-dimensional, nonlinear problems, a hybrid prediction model based on the random forest (RF) algorithm is developed in this paper. The model used the dragonfly algorithm (DA) to optimize two hyperparameters in RF, namely mtry and ntree, and used mean absolute error (MAE), root mean square error (RMSE), determination coefficient (R2), and variance accounted for (VAF) to evaluate model prediction performance. A database containing 217 sets of data was collected, with embedding depth (ED), drift span (DS), surrounding rock mass strength (RMS), joint index (JI) as input variables, and the excavation damaged zone thickness (EDZT) as output variable. In addition, four classic models, back propagation neural network (BPNN), extreme learning machine (ELM), radial basis function network (RBF), and RF were compared with the DA-RF model. The results showed that the DARF mold had the best prediction performance (training set: MAE = 0.1036, RMSE = 0.1514, R² = 0.9577, VAF = 94.2645; test set: MAE = 0.1115, RMSE = 0.1417, R² = 0.9423, VAF = 94.0836). The results of the sensitivity analysis showed that the relative importance of each input variable was DS, ED, RMS, and JI from low to high. [ABSTRACT FROM AUTHOR]

The paper addresses the limitations of the Moth-Flame Optimization (MFO) algorithm, a meta-heuristic used to solve optimization problems. The MFO algorithm, which employs moths’ transverse orientation navigation technique, has been used to generate solutions for such problems. However, the performance of MFO is dependent on the flame production and spiral search components, and the search mechanism could still be improved concerning the diversity of flames and the moths’ ability to find solutions. The authors propose a revised version called GMSMFO, which uses a Novel Gaussian mutation mechanism and shrink MFO to enhance population diversity and balance exploration and exploitation capabilities. The study evaluates the performance of GMSMFO using the CEC 2017 benchmark and 20 datasets, including a high-dimensional intrusion detection system dataset. The proposed algorithm is compared to other advanced metaheuristics, and its performance is evaluated using statistical tests such as Friedman and Wilcoxon rank-sum. The study shows that GMSMFO is highly competitive and frequently superior to other algorithms. It can identify the ideal feature subset, improving classification accuracy and reducing the number of features used. The main contribution of this research paper includes the improvement of the exploration/exploitation balance and the expansion of the local search. The ranging controller and Gaussian mutation enhance navigation and diversity. The research paper compares GMSMFO with traditional and advanced metaheuristic algorithms on 29 benchmarks and its application to binary feature selection on 20 benchmarks, including intrusion detection systems. The statistical tests (Wilcoxon rank-sum and Friedman) evaluate the performance of GMSMFO compared to other algorithms. The algorithm source code is available at https://github.com/MohammedQaraad/GMSMFO-algorithm. [ABSTRACT FROM AUTHOR]

Accurate forecasting of wind speed is crucial for power systems stability. Many machine learning models have been developed to forecast wind speed accurately. However, the accuracy of these models still needs more improvements to achieve more accurate results. In this paper, an optimized model is proposed for boosting the accuracy of the prediction accuracy of wind speed. The optimization is performed in terms of a new optimization algorithm based on dipper-throated optimization (DTO) and genetic algorithm (GA), which is referred to as (GADTO). The proposed optimization algorithm is used to optimize the bidrectional long short-term memory (BiLSTM) forecasting model parameters. To verify the effectiveness of the proposed methodology, a benchmark dataset freely available on Kaggle is employed in the conducted experiments. The dataset is first preprocessed to be prepared for further processing. In addition, feature selection is applied to select the significant features in the dataset using the binary version of the proposed GADTO algorithm. The selected features are utilized to learn the optimization algorithm to select the best configuration of the BiLSTM forecasting model. The optimized BiLSTM is used to predict the future values of the wind speed, and the resulting predictions are analyzed using a set of evaluation criteria. Moreover, a statistical test is performed to study the statistical difference of the proposed approach compared to other approaches in terms of the analysis of variance (ANOVA) and Wilcoxon signed-rank tests. The results of these tests confirmed the proposed approach’s statistical difference and its robustness in forecasting the wind speed with an average root mean square error (RMSE) of 0.00046, which outperforms the performance of the other recent methods.

In the field of safety detection of fruits and vegetables, how to conduct non-destructive detection of pesticide residues is still a pressing problem to be solved. In response to the high cost and destructive nature of existing chemical detection methods, this study explored the potential of identifying different pesticide residues on Hami melon by short-wave infrared (SWIR) (spectral range of 1000–2500 nm) hyperspectral imaging (HSI) technology combined with machine learning. Firstly, the classification effects of classical classification models, namely extreme learning machine (ELM), support vector machine (SVM), and partial least squares discriminant analysis (PLS-DA) on pesticide residues on Hami melon were compared, ELM was selected as the benchmark model for subsequent optimization. Then, the effects of different preprocessing treatments on ELM were compared and analyzed to determine the most suitable spectral preprocessing treatment. The ELM model optimized by Honey Badger Algorithm (HBA) with adaptive t-distribution mutation strategy (tHBA-ELM) was proposed to improve the detection accuracy for the detection of pesticide residues on Hami melon. The primitive HBA algorithm was optimized by using adaptive t-distribution, which improved the structure of the population and increased the convergence speed. Compared the classification results of tHBA-ELM with HBA-ELM and ELM model optimized by genetic algorithm (GA-ELM), the tHBA-ELM model can accurately identify whether there were pesticide residues and different types of pesticides. The accuracy, precision, sensitivity, and F1-score of the test set was 93.50%, 93.73%, 93.50%, and 0.9355, respectively. Metaheuristic optimization algorithms can improve the classification performance of classical machine learning classification models. Among all the models, the performance of tHBA-ELM was satisfactory. The results indicated that SWIR-HSI coupled with tHBA-ELM can be used for the non-destructive detection of pesticide residues on Hami melon, which provided the theoretical basis and technical reference for the detection of pesticide residues in other fruits and vegetables. [ABSTRACT FROM AUTHOR]

Due to the importance of the forest fire susceptibility zonation for proper management of this environmental hazard, this study presents two different hybrids of artificial neural network (ANN) for spatial analysis of forest fire in northern Iran. To this end, ant colony optimization (ACO) and biogeography-based optimization (BBO) evolutionary algorithms are synthesized with ANN to optimize its computational parameters. In this work, slope aspect, elevation, land use, wind speed, soil type, plan curvature, temperature, distance to river, distance from road, distance from village, slope degree, topographic wetness index, annual mean evaporation, annual mean rainfall, and normalized difference vegetation index are considered as the forest fire ignition factors. Notably, the frequency ratio model is used to demonstrate the spatial interaction between the forest fire and ignition factors. The findings showed that the BBO and ACO could improve the accuracy of the ANN from 81.3% to 84.0 and 83.9%, respectively. Moreover, the ranking results (obtained by applying mean square error, area under the curve, and mean absolute error indices) revealed the superiority of the BBO-ANN. [ABSTRACT FROM AUTHOR]

Estimation of carbon stock in seagrass meadows is in challenges of paucity of assessment and low accuracy of the estimates. In this study, we used a fusion of the synthetic aperture radar (SAR) Sentinel-1 (S-1), the multi-spectral Sentinel-2 (S-2), and coupled this with advanced machine learning (ML) models and meta-heuristic optimization to improve the estimation of total organic carbon (TOC) stock in the Zostera muelleri meadows in Tauranga Harbour, New Zealand. Five scenarios containing combinations of data, ML models (Random Forest, Extreme Gradient Boost, Rotation Forest, CatBoost) and optimization were developed and evaluated for TOC retrieval. Results indicate a fusion of S1, S2 images, a novel ML model CatBoost and the grey wolf optimization algorithm (the CB-GWO model) yielded the best prediction of seagrass TOC (R2, RMSE were 0.738 and 10.64 Mg C ha−1). Our results provide novel ideas of deriving a low-cost, scalable and reliable estimates of seagrass TOC globally. [ABSTRACT FROM AUTHOR]

The concept of mobile healthcare systems is promising. It is based on the cloud connected wireless biomedical wearables. In this scenario, the compression, processing, transmission and power effectiveness with precision are the key terms. A novel technique is presented for arrhythmia identification by processing the electrocardiogram signals. The solution is based on an effective hybridization of the multirate processing, QRS selection, variational mode decomposition, features mining from Modes, Metaheuristic optimization based features selection, and machine learning algorithms. The MIT-BIH dataset is used for experimentation. Performance of the Butterfly Optimization Algorithm, Manta Ray Foraging Optimization, and Emperor Penguin Optimization algorithms is investigated for features selection. A multi-subjects and multi-class dataset is used for testing the performance of classification by following the 10-fold cross validation strategy. The multirate processing with QRS selection and Metaheuristic optimization dependent features selection bring compression and aptitude for the processing and data transmission efficiencies. The system efficiently incorporates the multirate processing while securing an effective signal reconstruction. The respective compression gains and classification accuracies for the Butterfly Optimization Algorithm, Manta Ray Foraging Optimization, and Emperor Penguin Optimization algorithms are 27-fold, 29.45-fold & 46.29-fold and 99.14%, 99.08% & 98.65%. [ABSTRACT FROM AUTHOR]

Landslides and other catastrophic environmental disasters pose a significant danger to environmental, infrastructure, and people's lives. This research aimed to construct four optimized ensemble machine learning algorithms for landslide susceptibility (LS) mapping, namely particle swarm optimization (PSO) based artificial neural network (ANN), random forest, M5P, and support vector machine. The logistic regression (LR) model was then applied to the four-ensemble machine learning model and generated a hybrid optimized machine learning model. The receiver operating characteristics (ROC) curve was then used to validate LS map. The best model of four LS models depending on ROC's area under curve (AUC) is PSO-ANN (AUC-0.958) model. Also, LR model-based hybrid ensemble machine learning model achieved better accuracy (AUC: 0.962) than PSO-ANN model. Various resources, viz. grassland, built-up, and scarce foliage, are declared as landslide risk zones. Finally, elevation, soil-texture, slope, rainfall, and road distance are considered the most sensitive parameters for landslide occurrences. [ABSTRACT FROM AUTHOR]

With decreasing computational costs, improvement in algorithms, and the aggregation of large industrial and commercial datasets, machine learning is becoming a ubiquitous tool for process and business innovations. Machine learning is still lacking applications in the field of dynamic optimization for real-time control. This work presents a novel framework for performing constrained dynamic optimization using a recurrent neural network model combined with a metaheuristic optimizer. The framework is designed to augment an existing control system and is purely data-driven, like most industrial Model Predictive Control applications. Several recurrent neural network models are compared as well as several metaheuristic optimizers. Hyperparameters and optimizer parameters are tuned with parameter sweeps, and the resulting values are reported. The best parameters for each optimizer and model combination are demonstrated in closed-loop control of a dynamic simulation, and several recommendations are made for generalizing this framework to other systems. Up to 0.953% improvement is realized over the non-optimized case for a simulated coal-fired boiler. While this is not a large improvement in percentage, the total economic impact is $991,000 per year, and this study builds a foundation for future machine learning with dynamic optimization. [ABSTRACT FROM AUTHOR]

Background and aim: Machine learning methods are examined by many researchers to identify weeds in crop images captured by drones. However, metaheuristic optimization is rarely used in optimizing the machine learning models used in weed classification. Therefore, this research targets developing a new optimization algorithm that can be used to optimize machine learning models and ensemble models to boost the classification accuracy of weed images. Methodology: This work proposes a new approach for classifying weed and wheat images captured by a sprayer drone. The proposed approach is based on a voting classifier that consists of three base models, namely, neural networks (NNs), support vector machines (SVMs), and K-nearest neighbors (KNN). This voting classifier is optimized using a new optimization algorithm composed of a hybrid of sine cosine and grey wolf optimizers. The features used in training the voting classifier are extracted based on AlexNet through transfer learning. The significant features are selected from the extracted features using a new feature selection algorithm. Results: The accuracy, precision, recall, false positive rate, and kappa coefficient were employed to assess the performance of the proposed voting classifier. In addition, a statistical analysis is performed using the one-way analysis of variance (ANOVA), and Wilcoxon signed-rank tests to measure the stability and significance of the proposed approach. On the other hand, a sensitivity analysis is performed to study the behavior of the parameters of the proposed approach in achieving the recorded results. Experimental results confirmed the effectiveness and superiority of the proposed approach when compared to the other competing optimization methods. The achieved detection accuracy using the proposed optimized voting classifier is 97.70%, F-score is 98.60%, specificity is 95.20%, and sensitivity is 98.40%. Conclusion: The proposed approach is confirmed to achieve better classification accuracy and outperforms other competing approaches. [ABSTRACT FROM AUTHOR]

Cardiovascular diseases (CVD) are the most widely spread diseases all over the world among the common chronic diseases. CVD represents one of the main causes of morbidity and mortality. Therefore, it is vital to accurately detect the existence of heart diseases to help to save the patient life and prescribe a suitable treatment. The current evolution in artificial intelligence plays an important role in helping physicians diagnose different diseases. In the present work, a hybrid framework for the detection of heart diseases using medical voice records is suggested. A framework that consists of four layers, namely "Segmentation" Layer, "Features Extraction" Layer, "Learning and Optimization" Layer, and "Export and Statistics" Layer is proposed. In the first layer, a novel segmentation technique based on the segmentation of variable durations and directions (i.e., forward and backward) is suggested. Using the proposed technique, 11 datasets with 14,416 numerical features are generated. The second layer is responsible for feature extraction. Numerical and graphical features are extracted from the resulting datasets. In the third layer, numerical features are passed to 5 different Machine Learning (ML) algorithms, while graphical features are passed to 8 different Convolutional Neural Networks (CNN) with transfer learning to select the most suitable configurations. Grid Search and Aquila Optimizer (AO) are used to optimize the hyperparameters of ML and CNN configurations, respectively. In the last layer, the output of the proposed hybrid framework is validated using different performance metrics. The best-reported metrics are (1) 100% accuracy using ML algorithms including Extra Tree Classifier (ETC) and Random Forest Classifier (RFC) and (2) 99.17% accuracy using CNN. [ABSTRACT FROM AUTHOR]

It is a problematic task to perform petro-physical property prediction of carbonate reservoir rocks in most cases, specifically for permeability prediction since a carbonate rock most commonly contains grains of heterogeneous size distributions. Consequently, the permeability calculation of tight rocks in laboratories is costly and very time-consuming. Therefore, this study aims to tackle this issue by developing novel hybrid models based on combination of the modified version of the equilibrium optimizer (EO), i.e., MEO, and two conventional machine learning algorithms, namely extreme learning machine (ELM) and artificial neural network (ANN). The MEO employs a mutation mechanism in order to avoid trapping in local optima of EO by increasing the search capabilities. In this study, ELM-MEO and ANN-MEO, novel metaheuristic ELM-based and ANN-based algorithms, were constructed to predict the permeability of tight carbonates. In addition, ANN, ELM, RF, RVM and MARS combined with particle swarm optimization and genetic programming algorithm have a better insight into the performances for preferably predicting the permeability carbonates. The results illustrate that the proposed ELM-MEO model with R2 = 0.9323, RMSE = 0.0612 and MAE = 0.0442 in training stage and R2 = 0.8743, RMSE = 0.0806 and MAE = 0.0660 in testing stage, outperformed other ELM-based and ANN-based metaheuristic models in predicting the permeability of tight carbonates at all levels. [ABSTRACT FROM AUTHOR]

The next generation healthcare systems will be based on the cloud connected wireless biomedical wearables. The key performance indicators of such systems are the compression, computational efficiency, transmission and power effectiveness with precision. The electrocardiogram (ECG) signals processing based novel technique is presented for the diagnosis of arrhythmia. It employs a novel mix of the Level-Crossing Sampling (LCS), Enhanced Activity Selection (EAS) based QRS complex selection, multirate processing, Wavelet Decomposition (WD), Metaheuristic Optimization (MO), and machine learning. The MIT-BIH dataset is used for experimentation. Dataset contains 5 classes namely, "Atrial premature contraction", "premature ventricular contraction", "right bundle branch block", "left bundle branch block" and "normal sinus". For each class, 450 cardiac pulses are collected from 3 different subjects. The performance of Marine Predators Algorithm (MPA) and Artificial Butterfly Optimization Algorithm (ABOA) is investigated for features selection. The selected features sets are passed to classifiers that use machine learning for an automated diagnosis. The performance is tested by using multiple evaluation metrics while following the 10-fold cross validation (10-CV). The LCS and EAS results in a 4.04-times diminishing in the average count of collected samples. The multirate processing lead to a more than 7-times computational effectiveness over the conventional fix-rate counterparts. The respective dimension reduction ratios and classification accuracies, for the MPA and ABOA algorithms, are 29.59-times & 22.19-times and 98.38% & 98.86%. [ABSTRACT FROM AUTHOR]

To accurately estimate the rock shear strength parameters of cohesion (C) and friction angle (φ), triaxial tests must be carried out at different stress levels so that a failure envelope can be obtained to be linearized. However, this involves a higher budget and time requirements that are often unavailable at the early stage of a project. To address this problem, faster and more inexpensive indirect techniques such as artificial intelligence algorithms are under development. This paper first aims to utilize four machine learning techniques of Gaussian process regression (GPR), support vector regression (SVR), decision trees (DT), and long-short term memory (LSTM) to develop a predictive model to estimate parameters C and φ. To this aim, 244 datasets are available in the RockData software for intact Sandstone, including three input parameters of uniaxial compressive strength (UCS), uniaxial tensile strength (UTS), and confining stress (σ3) are employed in the models. The dropout technique is used to overcome the overfitting problem in LSTM-based models. A comprehensive evaluation is adopted for the performance indices of the prediction models. In this step, the most accurate results are produced by the LSTM model (C: R2 = 0.9842; RMSE = 1.295; MAPE = 0.009/φ: R2 = 0.8543; RMSE = 1.857; MAPE = 1.4301). In the second step, we improve the performance of the proposed LSTM model by fine-tuning the LSTM hyper-parameters, using six metaheuristic algorithms of grey wolf optimization (GWO), particle swarm optimization (PSO), social spider optimization (SSO), sine cosine algorithm (SCA), multiverse optimization (MVO), and moth flame optimization (MFO). The developed models' prediction performance for predicting parameter C from high to low was PSO-LSTM, GWO-LSTM, MVO-LSTM, MFO-LSTM, SCA-LSTM SSO-LSTM, and LSTM with ranking scores of 34, 29, 24, 21, 14, 12, and 5, respectively. Also, the models' prediction performance for predicting parameter φ from high to low was PSO-LSTM, GWO-LSTM, MVO-LSTM, MFO-LSTM, SCA-LSTM SSO-LSTM, and LSTM with ranking scores of 34, 31, 23, 18, 15, 14, and 5, respectively. However, the most robust results are produced by the PSO-LSTM model. Finally, the results indicate that applying a metaheuristic algorithm to tune the hyper-parameters of the LSTM model can significantly improve the prediction results. In the last step, the mutual information test method is applied to sensitivity analysis of the input parameters to predict parameters C and φ. Finally, it is revealed that parameters σ3 and UCS have the highest and lowest impact on the parameters C and φ, respectively. Highlights: Employing a large dataset consists of 244 data. Using six ML algorithms that most of them had not been tested before for this issue. Applying 5-fold CV to validate the results. Application of feature selection to find the most effective parameters on the water inflow into tunnels. Recognition of the best prediction method. [ABSTRACT FROM AUTHOR]

In the field of safety detection of fruits and vegetables, how to conduct non-destructive detection of pesticide residues is still a pressing problem to be solved. In response to the high cost and destructive nature of existing chemical detection methods, this study explored the potential of identifying different pesticide residues on Hami melon by short-wave infrared (SWIR) (spectral range of 1000–2500 nm) hyperspectral imaging (HSI) technology combined with machine learning. Firstly, the classification effects of classical classification models, namely extreme learning machine (ELM), support vector machine (SVM), and partial least squares discriminant analysis (PLS-DA) on pesticide residues on Hami melon were compared, ELM was selected as the benchmark model for subsequent optimization. Then, the effects of different preprocessing treatments on ELM were compared and analyzed to determine the most suitable spectral preprocessing treatment. The ELM model optimized by Honey Badger Algorithm (HBA) with adaptive t-distribution mutation strategy (tHBA-ELM) was proposed to improve the detection accuracy for the detection of pesticide residues on Hami melon. The primitive HBA algorithm was optimized by using adaptive t-distribution, which improved the structure of the population and increased the convergence speed. Compared the classification results of tHBA-ELM with HBA-ELM and ELM model optimized by genetic algorithm (GA-ELM), the tHBA-ELM model can accurately identify whether there were pesticide residues and different types of pesticides. The accuracy, precision, sensitivity, and F1-score of the test set was 93.50%, 93.73%, 93.50%, and 0.9355, respectively. Metaheuristic optimization algorithms can improve the classification performance of classical machine learning classification models. Among all the models, the performance of tHBA-ELM was satisfactory. The results indicated that SWIR-HSI coupled with tHBA-ELM can be used for the non-destructive detection of pesticide residues on Hami melon, which provided the theoretical basis and technical reference for the detection of pesticide residues in other fruits and vegetables.

As a frequent natural disaster, landslides incur significant economic and human losses worldwide. The main idea of this paper is to propose novel integrative models for landslide susceptibility evaluation in a prone area of Chaharmahal va Bakhtiari Province in Iran. To do this, four metaheuristic techniques, namely chimp optimization algorithm (ChOA), crow search algorithm (CSA), satin bowerbird optimization (SBO), and water cycle algorithm (WCA) are used to supervise the training of an artificial neural network (ANN). A spatial database is created by 170 historical landslides and 14 most common conditioning factors. The models are optimized in terms of hyper-parameters and elite ones are then used to produce the susceptibility maps. During the training phase, all models could acquire a reliable understanding of the landslide pattern. The obtained accuracy index of area under the receiving operating characteristic curve (AUROC) showed that the maps produced by the WCA-ANN (AUROC = 0.925) and SBO-ANN (AUROC = 0.900) are more accurate than those of the ChOA-ANN (AUROC = 0.851) and CSA-ANN (AUROC = 0.855). Considering this assessment, the suggested hybrid models are reliable enough for landslide susceptibility modeling. Accordingly, the generated susceptibility map can be used for urban planning and generating alert systems over the studied area. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) have evolved as a beneficial machine learning method that has been successfully used in various applications. Currently, DNN is a superior technique of extracting information from massive sets of data in a self-organized method. DNNs have different structures and parameters, which are usually produced for particular applications. Nevertheless, the training procedures of DNNs can be protracted depending on the given application and the size of the training set. Further, determining the most precise and practical structure of a deep learning method in a reasonable time is a possible problem related to this procedure. Meta-heuristics techniques, such as swarm intelligence (SI) and evolutionary computing (EC), represent optimization frames with specific theories and objective functions. These methods are adjustable and have been demonstrated their effectiveness in various applications; hence, they can optimize the DNNs models. This paper presents a comprehensive survey of the recent optimization methods (i.e., SI and EC) employed to enhance DNNs performance on various tasks. This paper also analyzes the importance of optimization methods in generating the optimal hyper-parameters and structures of DNNs in taking into consideration massive-scale data. Finally, several potential directions that still need improvements and open problems in evolutionary DNNs are identified. [ABSTRACT FROM AUTHOR]

Law enforcement plays an important role in road safety across the world. Speed enforcement merits special attention as the correlation between exceeding speed limits and the risk of fatalities and serious injuries is well-known. Methods based on radar and light detection and ranging techniques have established themselves as effective methods for measuring vehicle speeds in real time on roads. However, the price of instruments used for these measurements limits their widespread applicability. This work explored the potential of coupling artificial intelligence and audio analysis for vehicle speed assessment. The potential of deep neural networks and extreme gradient boosting is explored on a recently proposed real-world vehicle speed dataset. Frequency analysis techniques are also applied to help reduce the computational demands of experimentation. Additionally, a modified optimization algorithm is introduced to help select optimal control parameters and improve the performance of the suggested method. Five experiments were executed to demonstrate the potential for detecting the exact vehicle speed using regression techniques as well as utilizing classifications to detect vehicles breaking the speed limit. The proposed methods demonstrated promising potential when applied to this pressing challenge. • Introducing technique that uses audio analysis to detect exceeding speed limits. • Investigating capability of audio analysis in accurately estimating vehicle. • Vehicles exceed speed limit classification by optimized XGBoost. • Vehicle speed estimation by tuned deep neural network (DNN). • Introducing adapted metaheuristics to address XGboost and DNN tuning. [ABSTRACT FROM AUTHOR]

Multi-class classification is one of the major challenges in machine learning and an ongoing research issue. Classification algorithms are generally binary, but they must be extended to multi-class problems for real-world application. Multi-class classification is more complex than binary classification. In binary classification, only the decision boundaries of one class are to be known, whereas in multiclass classification, several boundaries are involved. The objective of this investigation is to propose a metaheuristic, optimized, multi-level classification learning system for forecasting in civil and construction engineering. The proposed system integrates the firefly algorithm (FA), metaheuristic intelligence, decomposition approaches, the one-against-one (OAO) method, and the least squares support vector machine (LSSVM). The enhanced FA automatically fine-tunes the hyperparameters of the LSSVM to construct an optimized LSSVM classification model. Ten benchmark functions are used to evaluate the performance of the enhanced optimization algorithm. Two binary-class datasets related to geotechnical engineering, concerning seismic bumps and soil liquefaction, are then used to clarify the application of the proposed system to binary problems. Further, this investigation uses multi-class cases in civil engineering and construction management to verify the effectiveness of the model in the diagnosis of faults in steel plates, quality of water in a reservoir, and determining urban land cover. The results reveal that the system predicts faults in steel plates with an accuracy of 91.085%, the quality of water in a reservoir with an accuracy of 93.650%, and urban land cover with an accuracy of 87.274%. To demonstrate the effectiveness of the proposed system, its predictive accuracy is compared with that of a non-optimized baseline model, single multi-class classification algorithms (sequential minimal optimization (SMO), the Multiclass Classifier, the Naïve Bayes, the library support vector machine (LibSVM) and logistic regression) and prior studies. The analytical results show that the proposed system is promising project analytics software to help decision makers solve multi-level classification problems in engineering applications. [ABSTRACT FROM AUTHOR]

Femoral neck fractures are a serious health problem, especially in the elderly population. Misdiagnosis leads to improper treatment and adversely affects the quality of life of the patients. On the other hand, when looking from the perspective of orthopedic surgeons, their workload increases during the pandemic, and the rates of correct diagnosis may decrease with fatigue. Therefore, it becomes essential to help healthcare professionals diagnose correctly and facilitate treatment planning. The main purpose of this study is to develop a framework to detect fractured femoral necks in PXRs (Pelvic X-ray, Pelvic Radiographs) while also researching how different machine learning approaches affect different data distributions. Conventional, LBP (Local Binary Patterns), and HOG (Histogram of Gradients) features were extracted manually from gray-level images to feed the canonical machine learning classifiers. Gray-level and three-channel images were used as inputs to extract the features automatically by CNNs (Convolutional Neural Network). LSTMs (Long Short-Term Memory) and BILSTMs (Bidirectional Long Short-Term Memory) were fed by automatically extracted features. Metaheuristic optimization algorithms, GA (Genetic Algorithm) and PSO (Particle Swarm Optimization), were utilized to optimize hyper-parameters such as the number of the feature maps and the size of the filters in the convolutional layers of the CNN architecture. The majority voting was applied to the results of the different classifiers. For the imbalanced dataset, the best performance was achieved by the 2-layer LSTM architecture that used features extracted from the fifth max-pooling layer of the CNN architecture optimized by GA. For the balanced dataset, the best performance was obtained by the CNN architecture optimized by PSO in terms of the Kappa evaluation metric. Although metaheuristic optimization algorithms such as GA and PSO do not guarantee the optimal solution, they can improve the performance on a not extremely imbalanced dataset especially in terms of sensitivity and Kappa evaluation metrics. On the other hand, for a balanced dataset, more reliable results can be obtained without using metaheuristic optimization algorithms but including them can result in an acceptable agreement in terms of the Kappa metric. [ABSTRACT FROM AUTHOR]

Multi-class classification is one of the major challenges in machine learning and an ongoing research issue. Classification algorithms are generally binary, but they must be extended to multi-class problems for real-world application. Multi-class classification is more complex than binary classification. In binary classification, only the decision boundaries of one class are to be known, whereas in multiclass classification, several boundaries are involved. The objective of this investigation is to propose a metaheuristic, optimized, multi-level classification learning system for forecasting in civil and construction engineering. The proposed system integrates the firefly algorithm (FA), metaheuristic intelligence, decomposition approaches, the one-against-one (OAO) method, and the least squares support vector machine (LSSVM). The enhanced FA automatically fine-tunes the hyperparameters of the LSSVM to construct an optimized LSSVM classification model. Ten benchmark functions are used to evaluate the performance of the enhanced optimization algorithm. Two binary-class datasets related to geotechnical engineering, concerning seismic bumps and soil liquefaction, are then used to clarify the application of the proposed system to binary problems. Further, this investigation uses multi-class cases in civil engineering and construction management to verify the effectiveness of the model in the diagnosis of faults in steel plates, quality of water in a reservoir, and determining urban land cover. The results reveal that the system predicts faults in steel plates with an accuracy of 91.085%, the quality of water in a reservoir with an accuracy of 93.650%, and urban land cover with an accuracy of 87.274%. To demonstrate the effectiveness of the proposed system, its predictive accuracy is compared with that of a non-optimized baseline model, single multi-class classification algorithms (sequential minimal optimization (SMO), the Multiclass Classifier, the Naïve Bayes, the library support vector machine (LibSVM) and logistic regression) and prior studies. The analytical results show that the proposed system is promising project analytics software to help decision makers solve multi-level classification problems in engineering applications.

Abstract Random single-hidden layer feedforward neural network (RSLFN) is currently a popular learning algorithm proposed for improving traditional gradient-based model due to its fast learning speed and acceptable performance. For RSLFN, the input weights and/or other parameters are randomly initialized, and the other ones are iteratively or non-iteratively trained. However, the performance of RSLFN is sensitive to the number of hidden neurons and randomly initialized parameters. Numerous methods have been successfully employed to improve the RSLFN from various perspectives. Because of their favourable search ability, metaheuristic optimization approaches gradually attract more and more attentions. Metaheuristic algorithms usually formulate the random parameters of RSLFN into an optimization model, and then provide a near-optimum solution which could be converted into RSLFN with better generalization performance. The hybrid method for optimizing RSLFN therefore shows considerable potential in intelligent computing and artificial intelligence. However, there is no comprehensive survey on RSLFN with metaheuristic in the research area, which ultimately leads to lost opportunities for an advancement. This paper firstly introduces the basic principles of RSLFN along with several metaheuristic algorithms. Secondly, it provides a comprehensive survey of the state-of-the-art contributions in the area. Finally, current challenges are highlighted and promising research directions are also presented. [ABSTRACT FROM AUTHOR]

• Novel route for cokeless methane to olefins, aromatics, and hydrogen (MTOAH). • Synergistic effect of hydrocarbon cofeeding on the MTOAH process was demonstrated. • Greater sustainability by metaheuristic optimization with machine learning models. • Successful prediction of the optimal reaction conditions for MTOAH. • Enhanced hydrocarbon selectivity in MTOAH, with suppressed coke formation. Methane to olefins, aromatics, and hydrogen (MTOAH) has received considerable attention because it can potentially provide an energy-efficient and environmentally friendly method for converting methane into valuable compounds. However, the optimization of the MTOAH process significantly hinders its economic feasibility because methane pyrolysis in the MTOAH process consumes considerable energy and carbon resources and requires several control factors. In this study, focusing on methane pyrolysis in the gas phase, a method for improving the MTOAH process with a hydrocarbon effect was successfully developed using metaheuristic optimization based on machine learning (ML) models. Because the trained ML models reasonably predicted various reaction outcomes, metaheuristic methods coupled with these ML models could optimize the reaction parameters to achieve the maximum hydrocarbon yield. By constructing score functions, various reaction conditions with maximum hydrocarbon yields and different coke selectivities were proposed. This provides key evidence for understanding the reaction mechanism and presents strategies for developing a sustainable cokeless catalytic reactor. Moreover, the importance of acetylene and the synergistic effect of hydrocarbon on the MTOAH process was verified using ML models. The optimal hydrocarbon content for cofeeding was determined based on the reaction temperature, but can be achieved at lower temperatures by loading a catalyst in a pre-gas phase reactor. Thus, our findings of the prioritization of the reaction parameters on the MTOAH mechanism will facilitate the development of a sustainable catalytic reactor under mild reaction conditions. [ABSTRACT FROM AUTHOR]

In this article, a series of learning strategies are proposed to enhance the optimization ability of the artificial electric field algorithm. Orthogonal learning is an important mathematical tool that can greatly influence the adaptability of population-based optimization algorithms. This article proposes, (i) an orthogonal array-based learning strategy to generate a better initial population for the artificial electric field algorithm. Along with the changes in the initialization mechanism, this article also proposes, (ii) an archive-based self-adaptive learning strategy for an artificial electric field algorithm. The proposed learning strategy divides the population into ordinary and extraordinary sub-populations, each with distinct learning mechanisms. The ordinary sub-population utilizes six learning strategies based on three archives, which contain individuals of different quality levels. We incorporate, (iii) a mutation strategy also to update the extraordinary sub-population. Finally, (iv) a self-adaptive strategy is implemented to dynamically adjust the parameters of the proposed algorithm. The effectiveness of these mechanisms is assessed through an extensive analysis of exploration–exploitation dynamics and diversity. Furthermore, an independent structural study is conducted to examine the impact of implemented mechanisms on the algorithm's behavior and efficiency. The proposed algorithm is evaluated on real parameter CEC 2017 problems across different dimensional search spaces. It is compared to eleven state-of-the-art algorithms, and the results demonstrate superior performance in terms of solution accuracy, convergence rate, search capability, and stability. The overall ranking highlights its exceptional potential for solving challenging optimization problems. Additionally, it outperforms other state-of-the-art algorithms across various dimensions, achieving accuracy rates of 64.48%, 70.05%, 78.73%, and 79.25% for dimensions 10, 30, 50, and 100, respectively. Furthermore, it demonstrates superior performance, outperforming others in 73.13% and 60.61% of the problems concerning average accuracy and statistical significance across all dimensions, respectively. [Display omitted] • Introducing an archive-based self-adaptive artificial electric field algorithm (AEFA) with orthogonal initialization for real-parameter optimization problems. • Proposal of an orthogonal array into the AEFA to initialize the population along with a self-adaptive parameter mechanism. • A novel strategy is proposed to divide the population into two homogeneous sub-populations. • Introduction of six learning strategies based on three elite archives and utilizing mutation mechanism. • A rigorous analysis of the proposed algorithm in terms of diversity, exploration–exploitation, convergence, and statistical validation. [ABSTRACT FROM AUTHOR]