Search

Machine learning (ML) and deep learning (DL) models are being increasingly employed for medical imagery analyses, with both approaches used to enhance the accuracy of classification/prediction in the diagnoses of various cancers, tumors and bloodborne diseases. To date however, no review of these techniques and their application(s) within the domain of white blood cell (WBC) classification in blood smear images has been undertaken, representing a notable knowledge gap with respect to model selection and comparison. Accordingly, the current study sought to comprehensively identify, explore and contrast ML and DL methods for classifying WBCs. Following development and implementation of a formalized review protocol, a cohort of 136 primary studies published between January 2006 and May 2023 were identified from the global literature, with the most widely used techniques and best-performing WBC classification methods subsequently ascertained. Studies derived from 26 countries, with highest numbers from high-income countries including the United States (n = 32) and The Netherlands (n = 26). While WBC classification was originally rooted in conventional ML, there has been a notable shift toward the use of DL, and particularly convolutional neural networks (CNN), with 54.4% of identified studies (n = 74) including the use of CNNs, and particularly in concurrence with larger datasets and bespoke features e.g., parallel data pre-processing, feature selection, and extraction. While some conventional ML models achieved up to 99% accuracy, accuracy was shown to decrease in concurrence with decreasing dataset size. Deep learning models exhibited improved performance for more extensive datasets and exhibited higher levels of accuracy in concurrence with increasingly large datasets. Availability of appropriate datasets remains a primary challenge, potentially resolvable using data augmentation techniques. Moreover, medical training of computer science researchers is recommended to improve current understanding of leucocyte structure and subsequent selection of appropriate classification models. Likewise, it is critical that future health professionals be made aware of the power, efficacy, precision and applicability of computer science, soft computing and artificial intelligence contributions to medicine, and particularly in areas like medical imaging. [ABSTRACT FROM AUTHOR]

This study presents a new intelligent computing method based on the Levenberg–Marquardt back-propagation-based supervised neural network (LMB-SNN) for the flow of Oldroyd 4-constant fluid in the forward roll coating process (OFC-FRCP). The mathematical formulations for OFC-FRCP are converted into ordinary differential equations using lubrication approximation theory (LAT) and suitable dimensionless parameters. The analytic expressions for velocity distribution, pressure and pressure gradients have been obtained using Adomian's Decomposition Method (ADM) for several situations. The fluid's separation point and coating thickness are then calculated using integration. The reference dataset for LMB-SNN has been obtained by ADM for multiple scenarios by adjusting various physical variables. The neural network's training, testing and validation are carried out in parallel to minimize the mean-squared error function (MSEF). The proposed LMB-SNN's effectiveness was confirmed by analyzing the MSEF, error histograms and regression plots. The precision of the method is verified by the close agreement between numerical outputs and dataset values. The intelligent numerical results computed using ADM and LMB-SNN are presented in various graphs and tables. The method's accuracy is demonstrated by the numerical outputs closest to the built and dataset values at levels between 1 0 − 9 and 1 0 − 1 2 . [ABSTRACT FROM AUTHOR]

The present study focused on the machinability of Ti-3Al-2.5 V for wire-electrical discharge machining (WEDM) using "BroncoCut-X wire" (zinc-coated copper wire). The machining characteristics have been evaluated by varying wire-tension (Tw), wire-speed (Sw), flushing-pressure (Pf), discharge current (Id), and spark-on-time (Son). The response characteristics associated with cutting-speed (Cs), kerf-width (KW), and surface roughness (RA) have been collected and analyzed using main-effect plots, scanning electron microscope (SEM), and analysis of variance (ANOVA). The maximum Cs and minimum KW and RA are obtained upto 8.90 mm/min, 3.34 µm and 0.2218 mm, respectively. Additionally, the novelty lies in the smart hybrid prediction tool considering the conflicting nature of responses are converted into single responses using Grey Relation Analysis (GRA) and Fuzzy Interference System (FIS) (Namely: Gray-Fuzzy Reasoning Grade (GFRG)). Furthermore, the optimal performance is calculated using Rao-algorithms (i.e., Rao1, Rao2, and Rao3). The obtained ideal machining condition is 16N wire-tension, 3 m/min wire-speed, 8 kg/mm2 flushing-pressure, 21A discharge current, and 14 µs spark-on-time. The result has also been compared with the JAYA-algorithm and improved-grey wolf optimizer (I-GWO) to demonstrate the efficacy of the intended approach. The confirmation test has been conducted and obtained that the GFRG-based results are further improved by using a hybrid GFRG-based Rao-algorithm of 9.55%, 2.36%, and 7.99% as Cs, KW and RA, respectively. Furthermore, this study shows that the proposed multi-objective optimization method not only leads to more stable solutions but also to shorter run times and enhanced quality to support engineers in reducing the cost of item failures. [ABSTRACT FROM AUTHOR]

Reservoir operations involve determining the quantity of water to be released or stored from the reservoir at any given time, based on the reservoir's current condition. This research presents optimizing techniques for reservoir operating policies on Ravishankar Sagar reservoir during monsoon season. Metaheuristics algorithms – GA, NSGA II, NSGA III, and Eps MOEA were used to optimize the objective function for maximization of storage along with a different set of constraints in Ravishankarsagar reservoir. Sensitivity analysis on all algorithms was performed to calibrate different evaluation parameters. The results were analyzed to find the effectiveness of soft computing algorithms in real-world problems. The standard deviation of 359.52, 351.93,349.68 and 361.09 and Median of 512.5, 525.36,568.96 and 433.79 was observed in GA, NSGA II, Eps MOEA, and NSGA IIII so all the algorithms performed well and were in close approximation, with the least standard deviation and the most optimum storage in Eps MOEA. [ABSTRACT FROM AUTHOR]

Electrical engineering models often rely on complex circuit configurations that facilitate the dynamic flow of electrically charged particles within a closed conductive network. These circuits serve as essential tools for simulating and analyzing diverse electrical systems and components. This paper introduces a study on nonlinear fractional circuits modeling through the development of a stochastic neuro-computational artificial intelligent-based solver to address mathematical models governing the Fractional order Caputo–Fabrizio stiff electric circuit model (FO-CFSECM) by manipulating the knacks of layered recurrent neural networks (LRNNs) trained with Gradient-based local search algorithm (GLA). In fractional calculus, the Caputo–Fabrizio (CF) fractional order derivative (FOD) emerges as a powerful instrument, binding its capabilities to deliver remarkably accurate solutions for fractional stiff systems. The objective of this work is to exploit the numerical treatment comprehensively for the dynamics of fractal Resistor–Capacitor (RC) and fractal Resistor–Inductor (RL) circuit models by introducing the CF fractional operator. Through the application of artificial intelligence-based soft computing and advanced back-propagative deep neural networks, a deeper understanding of the behavior and distinctive characteristics inherent in these models is sought. The Levenberg–Marquardt optimizer serves as an efficient training GLA tool for learning of LRNNs weights of fractal RL/RC circuit models. The comparative studies on variants of FO-CFSECM demonstrate that LRNNs achieve an impressive mean square error (MSE) ranging from 10−9 to 10−19 and absolute error (AE) within 10−6 to 10−8. The accuracy, reliability, and efficiency of LRNNs for solving the FO-CFSECM were further validated through MSE, AE, controlling parameters of state transitions, error histograms, and correlation measures. [ABSTRACT FROM AUTHOR]

The Storm Water Management Model (SWMM) is a reliable software program for simulating stormwater runoff in combined drainage facilities. However, it faces challenges in efficiency, especially when executing real-time forecasting. This research explores an alternative machine learning (ML) approach to predict water levels in sewer and street drainage systems utilizing SWMM data for model training. The goal is to construct a hybrid ML model for urban flooding prediction, considering hydrological conditions, geomorphologic properties, and drainage facility features. A robust training method is introduced to deliberate complex flow conditions. The suggested approach achieves favourable performance in predictive accuracy and computational efficiency. Findings include: (1) similarity between ML and SWMM flooding depth on street nodes increases from 62.3% to 96.2% with the proposed training; (2) the proposed model is about 50 times faster than SWMM in urban flood simulations; (3) reliable predictions from the ML model are demonstrated through four accuracy metrics. [ABSTRACT FROM AUTHOR]

This research aims to create a reliable model for estimating the compressive strength of cement mortar amended with metakaolin (MK) ingredients and predicting the impact of MK and the maximum diameter of the fine material (MDA) on the compressive strength of the mortar. This study collected 230 datasets from the research with various percentages and curing times. The water-to-binder ratio (w/b) ranges between 0.36 and 0.6 (by the weight of dry cement), sand to-binder ratio of 2–3, the metakaolin content of 0–30%, and the curing time is up to 90 days. Artificial neural network (ANN) models, nonlinear regression (NLR), multi-expression programming (MEP), and multivariate regression spline (MARS) models are the models. The coefficient of determination (R2), root-mean-squared error (RMSE), mean absolute error (MAE), scatter index (SI), standard deviation, and correlation coefficient between measured and predicted compressive strength are some of the evaluation tools that are used to measure the performance of the suggested models. The results show that the MARS model performs better with a high R2 and low RMSE and MAE than MEP, NLR, and ANN models. Based on the dispersion index, the MARS, MEP, and ANN, they have made good predictions about the compressive strength. According to the parametric analysis of MK and MDA, the ANN model effectively anticipated the impact of the previously mentioned model inputs and the ideal MK concentration for enhancing both long- and short-term compressive strength. [ABSTRACT FROM AUTHOR]

Concrete-filled steel tube columns (CFSTCs) are important elements in the construction sector and predictive analysis of their behavior is essential. Recent works have revealed the potential of metaheuristic-assisted approximators for this purpose. The main idea of this paper, therefore, is to introduce a novel integrative model for appraising the axial compression capacity (Pu) of CFSTCs. The proposed model represents an artificial neural network (ANN) supervised by satin bowerbird optimizer (SBO). In other words, this metaheuristic algorithm trains the ANN optimally to find the best contribution of input parameters to the Pu. In this sense, column length and the compressive strength of concrete, as well as the characteristics of the steel tube (i.e., diameter, thickness, yield stress, and ultimate stress), are considered input data. The prediction results are compared to five ANNs supervised by backtracking search algorithm (BSA), earthworm optimization algorithm (EWA), social spider algorithm (SOSA), salp swarm algorithm (SSA), and wind-driven optimization. Evaluating various accuracy indicators showed that the proposed model surpassed all of them in both learning and reproducing the Pu pattern. The obtained values of mean absolute percentage error of the SBO-ANN was 2.3082% versus 4.3821%, 17.4724%, 15.7898%, 4.2317%, and 3.6884% for the BSA-ANN, EWA-ANN, SOSA-ANN, SSA-ANN and WDA-ANN, respectively. The higher accuracy of the SBO-ANN against several hybrid models from earlier literature was also deduced. Moreover, the outcomes of principal component analysis on the dataset showed that the yield stress, diameter, and ultimate stress of the steel tube are the three most important factors in Pu prediction. A predictive formula is finally derived from the optimized SBO-ANN by extracting and organizing the weights and biases of the ANN. Owing to the accurate estimation shown by this model, the derived formula can reliably predict the Pu of concrete-filled steel tube columns. [ABSTRACT FROM AUTHOR]

Defect density is an essential software testing and maintenance aspect that determines the quality of software products. It is used as a management factor to distribute limited human resources successfully. The availability of public defect datasets facilitates building defect density prediction models using established static code metrics. Since the data gathered for software modules are often subject to uncertainty, it becomes difficult to deliver accurate and reliable predictions. To alleviate this issue, we propose a new prediction model that integrates gray system theory and fuzzy logic to handle the imprecision in software measurement. We propose a new similarity measure that combines the benefits of fuzzy logic and gray relational analysis. The proposed model was validated against defect density prediction models using public defect datasets. The defect density variable is frequently sparse because of the vast number of none‐defected modules in the datasets. Therefore, we also check our proposed model's performance against the sparsity level. The findings reveal that the developed model surpasses other defect density prediction models over the datasets with high and very high sparsity ratios. The ensemble learning techniques are competitive choices to the proposed model when the sparsity ratio is relatively small. On the other hand, the statistical regression models were the most inadequate methods for such problems and datasets. Finally, the proposed model was evaluated against different degrees of uncertainty using a sensitivity analysis procedure. The results showed that our model behaves stably under different degrees of uncertainty. [ABSTRACT FROM AUTHOR]

Test case prioritization (TCP) is a regression technique that sequences test cases by assigning priority based on specific criteria defined by software testers. Various parameters, such as code coverage, statement coverage, and method coverage, are utilized in Test Case Prioritization (TCP), wherein metaheuristic techniques are widely employed to determine the optimal order of test cases based on these specified parameters. However, simply applying these techniques does not ensure the satisfaction of all the needs of software testers. This paper introduces an empirical study that employs the multi-objective test case prioritization (MOTCP) technique to prioritize the test cases based on target points defined by software testers. The study calculates a Software Complexity Index (SCI) at the code level, identifying fault-prone areas. Furthermore, a Test-case Complexity Index (TCI) is also used for prioritization. The proposed technique incorporates various target points defined by the software tester to calculate SCI and TCI, which serve as our main objectives for TCP. A detailed analysis is also performed to examine the impact of these target points on the generated optimal order of test cases. Finally, the proposed model is compared with other state-of-the-art techniques across various evaluation parameters. [ABSTRACT FROM AUTHOR]

Purpose: This research is aimed at preparing landslide susceptibility using spatial analysis and soft computing machine learning techniques based on convolutional neural networks (CNNs), artificial neural networks (ANNs) and logistic regression (LR) models. Design/methodology/approach: Using the Geographical Information System (GIS), a spatial database including topographic, hydrologic, geological and landuse data is created for the study area. The data are randomly divided between a training set (70%), a validation (10%) and a test set (20%). Findings: The validation findings demonstrate that the CNN model (has an 89% success rate and an 84% prediction rate). The ANN model (with an 84% success rate and an 81% prediction rate) predicts landslides better than the LR model (with a success rate of 82% and a prediction rate of 79%). In comparison, the CNN proves to be more accurate than the logistic regression and is utilized for final susceptibility. Research limitations/implications: Land cover data and geological data are limited in largescale, making it challenging to develop accurate and comprehensive susceptibility maps. Practical implications: It helps to identify areas with a higher likelihood of experiencing landslides. This information is crucial for assessing the risk posed to human lives, infrastructure and properties in these areas. It allows authorities and stakeholders to prioritize risk management efforts and allocate resources more effectively. Social implications: The social implications of a landslide susceptibility map are profound, as it provides vital information for disaster preparedness, risk mitigation and landuse planning. Communities can utilize these maps to identify vulnerable areas, implement zoning regulations and develop evacuation plans, ultimately safeguarding lives and property. Additionally, access to such information promotes public awareness and education about landslide risks, fostering a proactive approach to disaster management. However, reliance solely on these maps may also create a false sense of security, necessitating continuous updates and integration with other risk assessment measures to ensure effective disaster resilience strategies are in place. Originality/value: Landslide susceptibility mapping provides a proactive approach to identifying areas at higher risk of landslides before any significant events occur. Researchers continually explore new data sources, modeling techniques and validation approaches, leading to a better understanding of landslide dynamics and susceptibility factors. [ABSTRACT FROM AUTHOR]

Accident prediction models are mathematical expressions or algorithms used to determine the causal factors for road accidents and road safety engineers are using these models, as well. Modelling this kind of accident is quite challenging and required good quality of data. The results of the artificial neural network model, Gaussian processes model, and support vector machine model are compared for vulnerable road accident frequency in this study. The accident frequency dataset comprises 218 records, with 146 designated for training purposes and 72 reserved for testing. The model's accuracy was contingent on: the mean absolute error, root mean square error and coefficient of correlation. The findings suggest that for predicting the vulnerable road user accidents on roads, the artificial neural network gives better correlation results as (0.912) that the support vector machine (0.879) and Gaussian processes (0.853). [ABSTRACT FROM AUTHOR]

The vertical deflection of steel–concrete composite bridges (VDCB) was estimated using novel ensemble soft computing (SC) models. These models, namely SGBE-RF, RSS-RF, and B-RF, are a combination of random forest and various ensemble techniques such as Stochastic Gradient Boosting (SGBE), random SUBSPACE (RSS), and bagging (B). Data from 83 bridges in Vietnam were obtained and utilized for the study. The importance of input variables used in prediction modeling was evaluated using correlation-based feature selection. The models were validated and compared using various methods including root mean squared error, mean absolute error, R-squared (R2), and Taylor diagrams. The validation results revealed that all three ensemble models, SGBE-RF (R2 = 0.805), RSS-RF (R2 = 0.781), and B-RF (R2 = 0.764) performed well in predicting the VDCB. Their performance surpassed that of the single RF model (R2 = 0.74). Among them, SGBE-RF emerged as the superior model. Therefore, it can be concluded that SGBE-RF is a powerful tool in accurate and quick prediction of the VDCB. [ABSTRACT FROM AUTHOR]

Energy piles are gaining popularity worldwide as both pile foundations and ground source heat producers. While considerable research has been conducted on energy pile design and evaluation, implementing these approaches effectively remains challenging for engineers. This study presents a novel approach based on Load and Resistance Factor Design (LRFD) with First-Order Second-Moment (FOSM) reliability analysis and target reliability ( β T ) for energy pile design. Additionally, three soft computing techniques, namely 1D-CNN, LSTM, and Bi-LSTM, were employed to enhance the design process. To conduct this research, datasets from each model were analyzed using various statistical indices, including R2, RMSE, RMSLE, MAPE, BF, AARD, NSE, U 95 , t-stat, TIC, and PI. The results indicated that the Bi-LSTM model demonstrated superior performance for heating operations, with R2, RMSE, and RMSLE values of 0.975, 0.021, and 0.007 for training and 0.984, 0.019, and 0.006 for testing datasets, respectively. For cooling operations, the LSTM model exhibited the most suitable predictive capabilities, achieving R2, RMSE, and RMSLE values of 0.969, 0.022, and 0.008 for training and 0.982, 0.019, and 0.007 for testing datasets, respectively. Comparative analyses, including rank analysis, Williams plot, accuracy matrix, and Taylor's diagram, were performed to assess the accuracy of the proposed models. The results demonstrated that the Bi-LSTM model for heating and the LSTM model for cooling outperformed other soft computing models within the reliability-based LRFD approach for energy pile design. This research contributes to the advancement of energy pile design methodologies, providing engineers with a more efficient and reliable LRFD-based framework. The successful application of soft computing techniques further enhances the accuracy of predictions for both heating and cooling operations. Overall, the findings of this study hold significant implications for the design and evaluation of energy piles, contributing to sustainable and energy-efficient construction practices. [ABSTRACT FROM AUTHOR]

Forecasting river flow is an important stage in reservoir operation, urban water management, and water resource optimization. The goal of this research is to forecast daily river flows for two intermittent and ephemeral rivers. Based on the antecedent river flow, the forecasting approach used stochastic (AR, ARIMA, and SARIMA) and machine learning (ML) techniques. The ML methods consist of three shallow learning models (NARnet, OS-ELM, and NAR-GMDH) and three deep learning models (LSTM, CNN-GRU, and stacked-LSTM). The precision of all the models in the ephemeral river was higher the intermittent river considering both base flow (R2ave = 0.94 vs. 0.87) and peak flow. The study was also extended to forecasting peak river flow values, demonstrating in the superiority of deep learning (RMSEave = 7.6 m3/s) over the shallow (RMSEave = 8.9 m3/s) and stochastic (RMSEave = 9.1 m3/s). Applied models acted similarly in forecasting peak flow in both rivers due to substantial variations in the floods. Moreover, the results demonstrated that the deep learning group models surpass stochastic and shallow learning group models in terms of five evaluation criteria, including RMSE, MAE, mean bias error, correlation coefficient, and agreement index for both rivers. In general, the results indicate that the CNN-GRU outperforms the other models in terms of river flow forecasting and is suggested as a viable model for learning the complicated behavior of streamflow. [ABSTRACT FROM AUTHOR]

This study aims to achieve the densest possible state of soil for constructing dams and roads. This requires assessing the compaction characteristics, Optimum Moisture Content (OMC), and Maximum Dry Density (MDD) to determine the soil's suitability for earthworks. However, this process is resource-intensive and time-consuming. To streamline the assessment, the study incorporates six parameters: gravel (G), sand (S), fine (F) contents, plastic limit (PL), liquid limit (LL), and plasticity index (PI). Four different models are used to predict compaction characteristics: artificial neural network (ANN), nonlinear regression (NLR), linear regression (LR), and multilinear regression (MLR). The study utilized a substantial dataset of 2162 entries, considering various soil gradation and plasticity properties as input variables. To evaluate the models' effectiveness, several statistical measures, including coefficient of determination (R2), scatter index (SI), root mean squared error (RMSE), mean absolute error (MAE), a20-index, and Objective (OBJ) value, were employed. The ANN model outperformed other models in predicting OMC, with RMSE, MAE, OBJ, SI, a20-index, and R2 values of 3.51, 2.31, 4.26, 0.202, 0.7, and 0.92%, respectively. However, for predicting MDD, the ANN model had the highest R2 value (R2 = 0.87), but the minimum RMSE (1.01), MAE (0.8), a20-index (0.998), and OBJ (1.07) were obtained from the MLR and LR models. Furthermore, sensitivity analyses revealed that the plastic limit significantly influences the OMC, while the gravel content plays a dominant role in predicting MDD. [ABSTRACT FROM AUTHOR]

Plant diseases are one of the factors that lead to yield and economic losses, which have a direct effect on national and international food production systems. One of the most essential ways to avoid agricultural product loss or reduction in amount is to diagnose plant diseases promptly and accurately. Hence, the diagnosis process for plants is crucial and should be conducted accurately. Moreover, this study focuses on this process by constructing an Artificiality Diagnostics Framework (ADF) to serve the study's objectives which entailed conducting diagnosis for plants in a professional and precise manner over uncertain environments. Thus, neutrosophic theory is considered the principal ingredient in our ADF. Due to the ability of neutrosophic to divide images into Truth (T), Falsity(F), and Indeterminacy (I). Also, deep learning (DL) is considered another principal ingredient in treating vast samples of datasets. Our comparative analysis of the leaves of potatoes is conducted whether leveraging neutrosophic and without utilizing Neutrosophic. ResNet50, ResNet152, and Mobile Net are the principal ingredients for the training dataset. The findings of implementing these networks indicated that ResNet50 achieved the highest accuracy of 0.915 in the T domain, ResNet152 achieved the highest accuracy of 0.905 in the True(T) domain, and Mobile Net achieved the highest accuracy of 0.915 in Truth(T) domain. Accuracy of 0.863 in Indeterminate(I). [ABSTRACT FROM AUTHOR]

The Internet of Things, or IoT, is a rapidly expanding technology, and one of its most important application areas is sustainable transportation. Choosing the right IoT service provider is a complex process that is considered a multi-criteria decision issue. The multi-criteria decision-making (MCDM) methodology deals with massive criteria. This study proposed an MCDM methodology for selecting the best IoT service provider under different sustainable criteria. The Multi-Objective Optimization Ratio Analysis (MOORA) method is an MCDM methodology used to rank the alternatives. The MOORA method is integrated with a type 2 neutrosophic set to deal with uncertain information. The proposed MCDM methodology applied to six criteria and 13 alternatives. The results show that alternative 3 is the best and alternative 5 is the worst. The sensitivity analysis was conducted to show the stability of the rank. The sensitivity analysis was performed under seven cases; the results show that the rank of alternatives was stable under different cases. The comparative study was conducted to show the effectiveness of the proposed methodology. [ABSTRACT FROM AUTHOR]

The domain adaptation (DA) transfer learning technique can accurately classify land cover in remote-sensing (RS) images, even with a small number of labeled samples. However, traditional architectures and various neural network and deep learning modifications make DA models difficult to learn quickly and computationally demanding. Extreme learning machine (ELM)-based DA models have shown potential for quick learning and improved generalization capabilities. Still, they are also known for their unstable and over-fitting characteristics and the ambiguity of the data sets. To address these difficulties, this article proposes two contributions. First, a modified fuzzy-rule-based ELM model (MFR-ELM) is created within a data-dependent platform to address the imprecision of the RS image land-cover classes and small sample size. Second, an ensemble of fuzzy rule-based ELM networks is proposed in the DA framework to address individual networks' instability and weak learning properties. The MFR-ELM and other fuzzy rule-based ELM designs are combined in the ensemble architecture for mutual benefit. The suggested model is proven to be superior to competing approaches for categorizing multispectral RS images, as supported by various performance assessment indices. [ABSTRACT FROM AUTHOR]

The majority of technological processes include dynamic systems, which are characterized by structural and parameter uncertainty. Conventional control procedures based on such models are unlikely to produce the desired performance since deterministic models cannot appropriately characterize these uncertainties. When dealing with complex nonlinear systems that are characterized by ill-defined and unknown elements, fuzzy-logic based system is an effective method. Fuzzy controllers—whose rule basis is built on the expertise of human experts—are frequently utilized in this context to achieve desired performance. This knowledge might not be adequate for some complex processes, thus different methods for creating IF-THEN rules have been suggested. In this paper, the design of higher-order fuzzy sets is discussed and a generalized novel algorithm is developed for implementing type-n fuzzy sets and it is applied for controlling the angle of steering for a vehicle in such a way that the vehicle tracks the desired path. In the proposed methodology, fuzzification and defuzzification are taking place n number of times, and defuzzification of higher order fuzzy values up to type-1 fuzzy set is performed using the center of gravity method. The obtained type-1 fuzzy values are used in the rule base and inference mechanism for calculating the final crisp output. The significance of the present research is that the computation cost is not as high as expected because both fuzzification and defuzzification processes are recursive and the reduced new type-1 fuzzy values have the additional ability to model the uncertainty and vagueness as compared to normal type-1 fuzzy values. A comparison of the controllers based on the type of fuzzy sets is done based on the square of error (SE). Both the disturbance signal and uncertainty scenarios were also considered. The use of higher-order fuzzy sets facilitates the controller to give the more accurate results for controlling the vehicle and the corresponding simulation results suggest the same. [ABSTRACT FROM AUTHOR]

Conventional methods for estimating porosity from core data are often limited by their spatial coverage, time-consuming nature, high cost, and inability to capture the entire underground reservoir. To address these challenges, this paper proposes a soft computing method using an Adaptive Neuro-Fuzzy Inference System (ANFIS) to estimate porosity in a conventional gas reservoir. The approach involves integrating well-logging data and the ANFIS model with a Particle Swarm Optimisation (PSO) training algorithm to predict the underground porosity model in the Hassi R’mel region of the Algerian Sahara. The choice of this hybrid method was based on its superior performance compared to other models. Although the Hassi R’mel reservoirs are of Triassic clay sandstones, originated by the fluviatile depositional environment that lay on top of the Hercynian surface, the characterisation of their properties still requires refinement to improve the reservoir performance and address the problems faced using appropriate technologies. With an average porosity of 15% and permeability ranging from 250 to 650 mD, the ANFIS method shows excellent accuracy compared to core data, and a reliability of 85%. Overall, the ANFIS-PSO hybrid model proves to be a dependable and efficient technique for porosity prediction, surpassing traditional methods. [ABSTRACT FROM AUTHOR]

Driven by global sustainability goals, understanding land surface temperature (LST) in urban thermal environments (UTE) is crucial. This study analyses remotely sensed LST data from Aqua and Terra satellites in a rapidly growing non-tier Indian city. It uniquely integrates remote sensing and soft computational analysis to predict daytime LST using artificial neural networks (ANN). The study is categorized based on transfer functions and hyper-parameters to enhance the comprehension of LST prediction and simulation scenarios by employing Back Propagation Neural Network (BPNN) model. Results indicate that the tansig transfer function yields the best performance overall in Aqua and Terra LST prediction, while purelin performs the least effectively. This study provides evidence that soft computing techniques effectively estimate LST data, offering a valuable solution for data scarcity issues. However, the findings are specific to the study area and may vary elsewhere. Exploring alternative methods and incorporating data from other areas could potentially improve accuracy and overall performance. Future research could focus on investigating the mechanisms behind different methods and their interactions with various factors. Additionally, integrating other variables like vegetation index, soil moisture, and others could contribute to a more holistic understanding of LST dynamics and enhance predictive capabilities. [ABSTRACT FROM AUTHOR]

Seismic risk assessment in bridge networks is of great interest in reducing earthquake adverse effects in structures and the society. Among existing methods for risk and loss assessment, Hazus methodology is widely accepted and used. The method includes several steps and calculations, being thus rather complex and time-consuming, especially if several bridges with diverse characteristics are to be examined. The present study investigates the potential of using soft computing techniques for determining the seismic hazard and expected losses as an alternative to the Hazus methodology. In particular, a number of representative datasets have been developed on the basis of Hazus and a number of methods have been formulated and tested regarding their capacity to effectively simulate the process. The methods that are examined range from regression curve-fitting to artificial neural networks of different size and characteristics. Alternative statistical tests have been used to evaluate the developed model capabilities. Evaluation results indicate that appropriately designed ANN methods can attain high accuracy and stable results among different datasets (training, testing, and validation). The analysis indicates the potential of using similar techniques to simulate seismic risk assessment based on actual data from earthquake events that have taken place in the past. [ABSTRACT FROM AUTHOR]

Dissolved oxygen (DO) is an important parameter in assessing water quality. The reduction in DO concentration is the result of eutrophication, which degrades the quality of water. Aeration is the best way to enhance the DO concentration. In the current study, the aeration efficiency (E20) of various numbers of circular jets in an open channel was experimentally investigated for different channel angle of inclination (θ), discharge (Q), number of jets (Jn), Froude number (Fr), and hydraulic radius of each jet (HRJn). The statistical results show that jets from 8 to 64 significantly provide aeration in the open channel. The aeration efficiency and input parameters are modelled into a linear relationship. Additionally, utilizing WEKA software, three soft computing models for predicting aeration efficiency were created with Artificial Neural Network (ANN), M5P, and Random Forest (RF). Performance evaluation results and box plot have shown that ANN is the outperforming model with correlation coefficient (CC) = 0.9823, mean absolute error (MAE) = 0.0098, and root mean square error (RMSE) = 0.0123 during the testing stage. In order to assess the influence of different input factors on the E20 of jets, a sensitivity analysis was conducted using the most effective model, i.e., ANN. The sensitivity analysis results indicate that the angle of inclination is the most influential input variable in predicting E20, followed by discharge and the number of jets. [ABSTRACT FROM AUTHOR]

In this paper, we first point out that the proof of the assertions (Theorems 3.12 and 3.25) in a previous paper by Jose and Mathew [Merin Jose and Sunil C. Mathew, Generalization of LM-filters: sum, subspace, product, quotient and stratification, Soft Computing 27(2023) 809-819] is incorrect. Then, we give the new constructs of the supremum and product of a family of LM-G-filters. Moreover, we give the definition of LM-G-fuzzy congruence relations on X. And, we prove that the category of LM-G-fuzzy congruence relational spaces and the category of LM-G-filter spaces are categorically isomorphic. [ABSTRACT FROM AUTHOR]

This research introduces a rule-based decision-making model to investigate corporate governance, which has garnered increasing attention within financial markets. However, the existing corporate governance model developed by the Security and Future Institute of Taiwan employs numerous indicators to assess listed stocks. The ultimate ranking hinges on the number of indicators a company meets, assuming independent relationships between these indicators, thereby failing to reveal contextual connections among them. This study proposes a hybrid rough set approach based on multiple rules induced from a decision table, aiming to overcome these constraints. Additionally, four sample companies from Taiwan undergo evaluation using this rule-based model, demonstrating consistent rankings with the official outcome. Moreover, the proposed approach offers a practical application for guiding improvement planning, providing a basis for determining improvement priorities. This research introduces a rule-based decision model comprising ten rules, revealing contextual relationships between indicators through if–then decision rules. This study, exemplified through a specific case, also provides insights into utilizing this model to strengthen corporate governance by identifying strategic improvement priorities. [ABSTRACT FROM AUTHOR]

In an era of data-based and information-centric Industry 4.0, extracting potential knowledge and valuable information from data is central to data mining tasks. Yet, the ambiguity, imprecision, incompleteness, and hybrid in real-world data pose tremendous challenges to critical information mining. Accordingly, we propose a new Max-Correlation Min-Redundant (MCMR) attribute reduction model from the uncertainty relation of attributes to avoid information loss in incomplete mixed data. Specifically, the neighbor relations are primarily developed based on the soft computing approach of the neighborhood information system, which divides the objects into neighborhood covers to maximize the utilization of the information in the incomplete mixed data. Then, we detailly analyze the internal and external consistency relationships of the four main uncertainty functions. Based on this, a new MCMR uncertain function is designed with maximum relevance and minimum redundancy. Experiments on nine real-world datasets validate the proposed model can improve data quality by mining critical information in classification tasks and achieving optimal performance with a minimum number of attributes. [ABSTRACT FROM AUTHOR]

Cyber-surveillance and connected devices can be misused to monitor, harass, isolate, and otherwise, harm individuals. In particular, these devices gather high volumes of personal data such as account details with shared passwords, person's behavior and preferences, movements by GPS, and audio-video recordings which can be maneuvered. It is therefore imperative to define approaches that help mitigate the Internet of things (IoT)-based real-time abuse in a pro-active, reactive, or predictive manner. The key objective of this research is to outline and categorize such approaches. Further, to comprehend predictive analytics as a potential solution to mitigate technology abuse, we propose an anomaly detection methodology (MFEW_Bagging) to classify normal and abnormal use pattern categories in an Intrusion Detection System (IDS) for IoT system. A hybrid feature selection technique based on an ensemble of multiple filter–based techniques and a wrapper algorithm is firstly used as search method for finding an optimal feature subset. Further, ensemble learning technique, namely bagging, is used for final classification into normal and abnormal use pattern categories. The use of ensemble feature selection removes biasness of individual feature selection method during ensemble and identifies the optimal subset with non-redundant and relevant features. The proposed methodology is evaluated on publicly available real-time IDS dataset. The research persuades the need of designing robust and lightweight IDS for IoT-based smart environments which understand the cyber-security risks in a proactive predictive manner as it the best way to defend networks and systems with the growing IoT complexity. [ABSTRACT FROM AUTHOR]

In software engineering, the planning and budgeting stages of a software project are of great importance to all stakeholders, including project managers as well as clients. The estimated costs and scheduling time needed to develop any software project before and/or during startup form the basis of a project's success. The main objective of soft- ware estimation techniques is to determine the actual effort and/or time required for project development. The use of machine learning methods to address the estimation problem has, in general, proven remarkably successful for many engineering problems. In this study, a fully connected neural network (FCNN) model and a metaheuristic, gray wolf optimizer (GWO), called GWO-FC, is proposed to tackle the software development effort estimation (SEE) problem. The GWO is integrated with FCNN to optimize the FCNN parameters in order to enhance the accuracy of the obtained results by improving the FCNN's ability to explore the parameter search field and avoid falling into local optima. The proposed technique was evaluated utilizing various benchmark SEE datasets. Furthermore, various recent algorithms from the literature were employed to verify the GWO-FC performance. In terms of accuracy, comparative outcomes reveal that the GWO-FC performs better than other methods in most datasets and evaluation criteria. Experimental outcomes reveal the strong potential of the GWO-FC method to achieve reliable estimation results. [ABSTRACT FROM AUTHOR]

In the present study, various soft computing techniques were employed for predicting the Marshall stability (MS) of asphalt concrete reinforced with polypropylene fibre, namely Multiple Linear Regression (MLR), Reduced Error Pruning (REP), Random tree (RT), Random Forest (RF), M5P_pruned and M5P_unpruned-based models, and the performance of these models was evaluated statistically. A total data set of 138 samples has been collected from various reliable and authentic published experimental researches, out of which 96 samples are used for training and 42 samples for testing using 70:30 split ratio. For each soft computing technique, coefficient of correlation (CC), Wilmott index (WI), mean absolute error (MAE), root-mean-square error (RMSE), Nash–Sutcliffe model efficiency coefficient (NSE), scattering index (SI) and mean absolute percentage error (MAPE) were calculated as the statistical indicators for analysing the performance of each soft computing technique. The RF model's dominance over the other models was confirmed by statistical indicators with values of CC as 0.9141, WI as 0.950, MAE as 1.1028, RMSE as 1.661, NSE as 1.0006, SI as 0.12077 and MAPE as 7.802, while the RT model also has shown competitive prediction ability over the M5P_pruned and M5P_unpruned, REP and MLR models. A sensitivity analysis presented that the bitumen content was the most effective parameter for predicting the Marshall stability using RF-based model. The results of this computational evaluation clearly showed that the applied soft-computing technique was capable of accurately calculating the Marshall stability of the asphalt concrete. [ABSTRACT FROM AUTHOR]

In this paper, the strength of basalt fiber reinforced concrete (BFRC) was investigated. Four percentages of basalt fiber (0.1%, 0.2%, 0.3%, and 0.4%) by volume fraction were added to concrete mix and BFRC cubes of 150 mm size were prepared and tested after 28 days curing age. It was found that the compressive strength of concrete with basalt fibers (1–4 percent) decreased in the range of 9.71–18.62 percent for M25. To predict the compressive strength of BFRC, three soft computing techniques, i.e., Random Forest (RF), Stochastic Random Tree, Artificial Neural Network (ANN) were applied for Nine inputs, i.e., Cement, Fine Aggregate, Coarse Aggregate, Water, Fly Ash, Superplasticizer, BF, length of Fiber, curing time, and which is Compressive Strength as output. The results showed correlation coefficient (CC) were (0.9938, 0.9755), (0.9985, 0.9342), and (0.9793, 0.7308) in training and testing stages, respectively, for RF, Stochastic RT and ANN. The Stochastic Random Tree was the outperforming model among the three applied ones. The sensitivity analysis showed that curing period is the most critical input in compressive strength. [ABSTRACT FROM AUTHOR]

An unsteady two-dimensional magnetized Casson nanofluid flow model is constructed over a wedge under the effect of thermal radiation and chemical reaction. The multiple slip effects are also assumed near the surface of the wedge along with the convective boundary restrictions. This study investigates the application of soft computing techniques to address the challenges posed by the complexity of problem modeling and numerical methods. Traditional approaches incorporating various model factors may struggle to provide accurate solutions. To resolve this issue, Gaussian process regression (GPR) is employed to predict the solution of the proposed flow model. With the help of the numerical shooting method together with Runge–Kutta–Fehlberg fourth-fifth-order (RKF-45) reference data, the GPR model is trained. The numerical simulation illustrated that the Casson fluid parameter β and the unsteadiness parameter S strengthen the friction factor, and the heat transfer rate is enhanced as the radiation parameter R d becomes larger. In addition, the Biot numbers B i 1 & B i 2 lead to strengthen nanoparticle temperature; an opposite behavior is noticed with the skin friction coefficient S ˜ f x R e x 0.5 , heat transfer rate H ˜ t x R e x 0.5 , and nanoparticle transfer rate C ˜ t x R e x 0.5 . The GPR model with the exponential Kernel function provided better performance than other functions on both training and checking datasets to predict S ˜ f x R e x 0.5 , H ˜ t x R e x 0.5 , and C ˜ t x R e x 0.5 . Statistical metrics including RMSE, MAE, MAPE, MSE, R 2 , and R are employed to check the accuracy and convergences of the predicted and numerical solutions obtained through GPR and RKF-45. It is observed that all three GPR models had an R 2 value of higher than 0.9. The proposed study demonstrates the advantages of employing soft computing methods (GPR) to effectively analyse the behavior of complex flow models. [ABSTRACT FROM AUTHOR]

AbstractA robust, accurate and reliable approach is essential for not only examining people’s acceptance of cloud-based learning management systems in developing nations but also for accurate prediction of factors affecting its implementation and progress. Therefore, in this study, three different soft computing models; Adaptive neuro-fuzzy inference system (ANFIS), Support vector regression (SVR), and Emotional artificial neural network (EANN), were employed for predictions of factors affecting cloud-based learning management systems take-up and progress using data gotten from six Nigerian colleges. The performance of the models was assessed using five arithmetic metrics; MAPE, NSE, RMSE, rRMSE, and RM. All the proposed models forecast the effects of the study inputs on LMS with higher accuracy (NSE > 0.98). However, the SVR model outshone the other models as it increased the performance of the study-reported model by 2% and 4% respectively. Based on the study results, instructors’ quality, motivation, and resource availability were found to be the key factors that affect cloud-based learning technologies take-up and progress in the study area. Interestingly, unlike prior studies, this study found system ease of use and usefulness to have insignificant effects on LMS take-up. Finally, the practical implications and limitations of the study were discussed based on the study findings. [ABSTRACT FROM AUTHOR]

Cognitive maps are a vital tool that can be used for knowledge representation and reasoning. Fuzzy Cognitive Maps (FCMs) are popular soft computing techniques used to model large and complex systems, and they can aid in explainable artificial intelligence (AI). FCMs, however, cannot model the indeterminacy that arises in a system due to various uncertainties. Neutrosophic Cognitive Maps (NCMs), upgraded FCMs that could model indeterminacy, were introduced to address this issue. NCMs are a generalization of FCMs, a field of cognitive science firmly based on neural networks. NCMs have been used to solve a wide range of problems. NCMs were introduced in 2002, and even after 20 years, NCMs do not have any supportive software, package, toolbox, or visualization software like FCMs. The main reason for the absence of dedicated software is due to the indeterminacy concept 'I' and how it has to be handled. This paper presents the dedicated Python package created for handling the functioning of NCMs. The modelling software presented in this paper aids in visualizing the NCMs as a signed digraph with indeterminacy that is a directed signed neutrosophic graph. This package implements a sample case study using NCMs. [ABSTRACT FROM AUTHOR]

In this paper, two controller design techniques such as linear active disturbance rejection control (LADRC) and generalized active disturbance rejection control (GADRC) are proposed for solar tracker system. The merit of the proposed method is that there is no need of tuning of proportional-integral-derivative (PID) parameters which is normally done in existing PID controller design techniques. As, the soft computing PID-based techniques involve very large iterations, so they consume much time. Also, these techniques are computationally inefficient. Therefore, to get rid of the conventional PID controller design topology, the LADRC scheme is proposed, which requires very few information of the system and even then the performance of the controller is impressive. The performance of the controller is further improved via GADRC scheme by including complete information of the system. Next, the robustness of proposed methods is investigated by the presence of system parametric uncertainty. Further, the performance of the proposed controllers is checked under time delays. Simulation results reveal that the proposed methods exhibit superb capability of disturbance rejection, high robustness and excellent transient performance in the presence of time delay in comparison to numerous existing PID-based control techniques. [ABSTRACT FROM AUTHOR]

This work centers on methyl ester yield modeling; by Azadirachta Indica seed oil (AISO) transesterification, using Adaptive Neuro-fuzzy Inference System (ANFIS), Artificial Neural Network (ANN) and Response Surface Methodology (RSM). AISO was obtained from the seeds of Azadirachta Indica tree. The oils were extracted from the seeds using solvent extraction method. The physicochemical properties of AISO and Azadirachta Indica seed oil methyl ester (MAISOt ) were determined using standard methods. Fatty acid composition was determined using, Gas Chromatography (GC). Statistical evaluations of these models show their efficacy in the order RSM < ANN < ANFIS, with ANFIS as the best; as indicated by its very high R² value of 0.9999 and low RMS error value of 0.0011. The ANFIS predicted minimum and maximum values for % methyl ester yields were 54.66 and 90.25 %, respectively, while the other models predicted similar methyl ester yields. The physicochemical characterization results of AISO and MAISOt samples, show that their respective viscosity, dielectric strength (DS), pour and flash points values were (8.83 and 3.47 mm² s−1 ), (33.42 and 48.93 KV), (9 and -6 ◦C), and (162 and 174 ◦C). These results indicated the MAISOt sample’s potential use as a transformer fluid. GC result indicated that MAISOt was unsaturated. Finally, on the basis of the gotten model results, ANFIS was adjudged as the best predictive model, followed by ANN and RSM, in that order [ABSTRACT FROM AUTHOR]

This work aims to create an ANN-based system for a musical improviser. An artificial improviser of "hearing" music will create a melody. The data supplied to the improviser is MIDI-type musical data. This is the harmonic-rhythmic course, the background for improvisation, and the previously made melody notes. The harmonic run is fed into the system as the currently ongoing chord and the time to the next chord, while the supplied few dozen notes performed earlier will indirectly carry information about the entire run and the musical context and style. Improvisation training is carried out to check ANN as a correct-looking musical improvisation device. The improviser generates several hundred notes to be substituted for a looped rhythmic-harmonic waveform and examined for quality. [ABSTRACT FROM AUTHOR]

The presence of malicious software (malware), for example, in Android applications (apps), has harmful or irreparable consequences to the user and/or the device. Despite the protections app stores provide to avoid malware, it keeps growing in sophistication and diffusion. In this paper, we explore the use of machine learning (ML) techniques to detect malware in Android apps. The focus is on the study of different data pre-processing, dimensionality reduction, and classification techniques, assessing the generalization ability of the learned models using public domain datasets and specifically developed apps. We find that the classifiers that achieve better performance for this task are support vector machines (SVM) and random forests (RF). We emphasize the use of feature selection (FS) techniques to reduce the data dimensionality and to identify the most relevant features in Android malware classification, leading to explainability on this task. Our approach can identify the most relevant features to classify an app as malware. Namely, we conclude that permissions play a prominent role in Android malware detection. The proposed approach reduces the data dimensionality while achieving high accuracy in identifying malware in Android apps. [ABSTRACT FROM AUTHOR]

Nanofluids are thermoelectric substance with a greater thermal transmission effectiveness than traditional thermal fluids. Nanofluid's specific heat capacity (SHC) is a crucial thermophysical parameter since it controls the fluid's heat transfer coefficient. Surprisingly few research investigations have been done on the prognostic modelling of the specific heat capacity of fusion nanofluids, despite the fact that numerous experiments have been conducted on the heat transfer capacitance of blended nanofluids. This study focuses on the use of algorithms based on machine learning procedures (MLP) to estimate the SHC of blended nanofluids. Numerous numbers of hybrid nanofluids are investigated in this investigation. Total of 984 samples of hybrid nanofluids for specific heat capacity has been collected from nine experimental research papers. Various MLP techniques were utilized in this analysis, including extreme boost gradient boost regression (XGB), support vector regression augmented with a genetic algorithm (support vector regression [SVR]‐genetic algorithm [GA]), grid search optimized based gradient boost regression algorithm (GBR) (GBR‐GSO), and voting ensemble procedure (VE). The obtained correlation coefficients of the SVR‐GA, XGB, VE and GBR‐GSO models for the testing dataset are 98.43%, 96.29%, 94.4% and 96.55% respectively. The SVR‐GA model showed a better predictive accuracy relative to other ML models. This SVR‐GA anticipated model could be used for quick and reliable prediction of the SHC of blended nanofluids which reduces the burden associated with experimental measurement possible future work includes applying a machine‐learning strategy to the problem of determining the diffusivity of hybrid nanofluids. [ABSTRACT FROM AUTHOR]

Vortex-assisted dispersive solid-phase extraction technique was used as a rapid and efficient method for preconcentration of ultra-trace levels of total thallium (Tl) followed by electrothermal atomic absorption spectrometry. Graphene oxide-modified polyvinylpyrrolidone nanocomposite was prepared as an adsorbent and characterised by the Fourier transform infrared spectrometry, field emission-scanning electron microscopy, energy-dispersive X-ray spectroscopy , and X-ray diffraction spectroscopy. To find the optimum conditions for extraction of ultra-trace levels of Tl (ӀӀӀ), the response surface methodology based on central composite design was used. Based on the results, pH = 6.6, amounts of adsorbent = 7.7 mg, extraction time = 27 min and desorption time = 5 min provide maximum extraction recovery for Tl (ӀӀӀ). Under the optimum conditions, the calibration curve was linear in the range of 0.08–1.5 µg L−1 Tl (ӀӀӀ) with the R2 value of 0.9985. The relative standard deviation was 4.0% (n = 7) and the detection limit was 0.019 µg L−1 (n = 8). Also, the enrichment factor which could be calculated from the slope of the calibration curve after preconcentration step to that without preconcentration was 95.8. To understand the adsorption mechanism, two-parameter and three-parameter adsorption isotherms were studied, and the obtained results show that the adsorption of Tl (ӀӀӀ) followed by the Freundlich isotherm and the maximum adsorption capacity was 142.8 mg g−1. Also, the results of adsorption kinetic show that the adsorption of Tl (ӀӀӀ) was followed by the pseudo-first-order kinetic model. Moreover, random tree (RT) and artificial neural network (ANN) are employed for prediction of adsorption performance based on effective parameters. The outcomes of soft computing demonstrated that RT (R2 = 0.95) and ANN (R2 = 0.87) have acceptable accuracy for estimation and prediction of extraction recovery of Tl (ӀӀӀ). [ABSTRACT FROM AUTHOR]

Recently, image thresholding methods based on various entropy functions have been found popularity. Nonetheless, entropic-based methods depend on the spatial distribution of the grey level values in an image. Hence, the accuracy of these methods is limited due to the non-uniform distribution of the grey values. Further, the analysis of the COVID-19 X-ray images is evolved as an important area of research. Therefore, it is needed to develop an efficient method for the segmentation of the COVID-19 X-ray images. To address these issues, an efficient non-entropy-based thresholding method is suggested. A novel fitness function in terms of the segmentation score (SS) is introduced, which is used to reduce the segmentation error. A soft computing approach is suggested. An efficient optimizer using the chance-based birds' intelligence is introduced to maximize the fitness values. The new optimizer is validated utilizing the benchmark test functions. The statistical parameters reveal that the suggested optimizer is efficient. It shows a quite significant improvement over its counterparts—optimization based on seagull/cuckoo birds. Precisely, the paper includes three novel contributions—(i) fitness function, (ii) chance-based birds' intelligence for optimization, (iii) multiclass segmentation. The COVID-19 X-ray images are taken from the Kaggle Radiography database, to the experiment. Its results are compared with three different state-of-the-art entropy-based techniques—Tsallis, Kapur's, and Masi. For providing a statistical analysis, Friedman's mean rank test is conducted and our method Ranked one. Its superiority is claimed in terms of Peak Signal to Noise Ratio (PSNR), Feature Similarity Index (FSIM) and Structure Similarity Index (SSIM). On the whole, an improvement of about 11% in PSNR values is achieved using the proposed method. This method would be helpful for medical image analysis. [ABSTRACT FROM AUTHOR]

This research aims to conduct a comparative analysis of the first crack load, flexural strength, and shear strength in reinforced concrete beams without stirrups. The comparison is made between the conventional model developed according to the current design code (ACI building code) and an unconventional approach using Artificial Neural Networks (ANNs). To accomplish this, a dataset comprising 110 samples of reinforced concrete beams without stirrup reinforcement was collected and utilised to train a Multilayer Backpropagation Neural Network in MATLAB. The primary objective of this work is to establish a knowledge-based structural analysis model capable of accurately predicting the responses of reinforced concrete structures. The coefficient of determination obtained from this comparison yields values of 0.9404 for the first cracking load, 0.9756 for flexural strength, and 0.9787 for shear strength. Through an assessment of the coefficient of determination and linear regression coefficients, it becomes evident that the ANN model produces results that closely align with those obtained from the conventional model. This demonstrates the ANN's potential for precise prediction of the structural behaviour of reinforced concrete beams. [ABSTRACT FROM AUTHOR]

The extensive utility of sensor technology and M2M networking are the future of industries for the production of commercial and non-commercial applications. One such application, the Internet of Things (IoT), was the universally accepted technology of sensors and wide-area networking. This causes severe effects of low power quality and results in the device malfunctioning. The increase in the use of smart devices will be a future challenge for power quality and its effect on associative devices. The main contribution of this survey is to analyze harmonic distortions due to non-linear load characteristics in the power management unit of traditional systems & advanced smart IoT devices. It helps to understand the utility of conventional harmonic mitigation techniques and the need for adaptive features for the latest technology. The conventional compensation techniques are incorporated with machine learning approaches to deal with the uncertainty of load variation on real-time signals. This survey gives a brief insight into the present scope of conventional compensation techniques and futuristic soft computing techniques to mitigate harmonic distortions. Additionally, the survey is intended to call the attention of the researchers to the future perspective of smart devices and to make the academic community aware of the existing practices. [ABSTRACT FROM AUTHOR]