Artificial Neural Network Analysis Research Articles

Understanding Isoniazid Crystals Agglomeration Through Quality by Design and Artificial Neural Network Analysis

Introduction: The aims of this study were to determine and govern critical sources of variability in the process and to investigate the effect of excipients and processing conditions on the critical quality attributes (CQAs) of isoniazid (INZ) crystal agglomerates using a quality-bydesign (QbD) paradigm along with risk-based approach. Methods: The QbD paradigm is thoroughly exercised to determine the CQAs and to, link these CQAs to agglomerate properties and to recognize potentially significant input variables. Considering this, a risk assessment tool was executed with different excipients and processing conditions to define their effect on CQAs of agglomerates. Potential risk factors were recognized using an Ishikawa diagram and then investigated using principal component analysis and an artificial neural network. After that, all agglomerates were optimized using the design of the experiment. Results: The optimized INZ crystal agglomerates were characterized using various tools such as X-ray diffractometry, Fourier transform infrared spectroscopy, scanning electron microscopy, differential scanning calorimetry, gas chromatography, and stability testing. These characterizations confirmed that INZ did not undergo any structural modifications in the agglomerates. Conclusion: The results indicate that the QbD principles and risk assessment tool provide a valuable means to fully understand directly compressible INZ crystal agglomerates, which could be considered an exceptional alternative to the granulation method.

An SEM-ANN analysis of the influence of strategic merger on strategic postchange of family businesses

Purpose Companies seek to increase the percentage of acquisitions in different parts of the world by expanding operations. Many companies are adopting strategic mergers to expand their influence. However, most strategic change programs fail to achieve their objectives. This study aims to investigate employees’ reactions after strategic mergers through the mediating role of the employees’ psychological context. It was necessary to identify the most prominent postmerger employees’ behaviors. The study addressed this gap by investigating the outcomes of strategic mergers. Design/methodology/approach Data for this study were collected from 30 family businesses. Accordingly, 341 questionnaires were collected with an overall response rate of 64%. The structural equation modeling (PLS-SEM) approach and the nonlinear relationships approach were adopted by implementing artificial neural network (ANN) analysis. Findings The results confirm that there is a clear impact of strategic mergers on employees’ postmerger behavior because of the change at the hierarchical level and the process of distributing roles. Employees’ psychological context (individual incentives, anxiety and individual mobbing) mediates the relationship between strategic mergers and postmerger employees’ behavior. In addition, individual incentives are considered the main contributor to retaining or not retaining employees in family businesses after strategic merger. Research limitations/implications Policymakers in organizations must pay attention to employees’ possible reactions to the internal and external policies of the organization by increasing individual incentives and reducing individual mobbing toward strategic merger. This study has theoretical implications that are critical guidelines for academics in mitigating the negative consequences for employees’ postmerger behavior. This study captured linear and nonlinear relationships to discover the determinants and antecedents of a strategic merger in family businesses. However, future studies should focus on using more robust statistical methods by adopting decision-making methods to determine the best and worst companies in terms of adopting strategic mergers. Originality/value The scarcity of literature on the most important determinants of postmerger employees’ behavior is considered an encouragement to conduct the current study. To this end, this study enriches the ongoing and future literature by examining the most important factors influencing the strategic merger of family businesses. Family businesses have changed the economic landscape of many countries. The investigation of the strategic merger of these companies is considered a worthy matter of study to improve the nation’s economy.

Application of artificial neural network for the mechano-bactericidal effect of bioinspired nanopatterned surfaces.

This study aimed to calculate the effect of nanopatterns' peak sharpness, width, and spacing parameters on P. aeruginosa and S. aureus cell walls by artificial neural network and finite element analysis. Elastic and creep deformation models of bacteria were developed in silico. Maximum deformation, maximum stress, and maximum strain values of the cell walls were calculated. According to the results, while the spacing of the nanopatterns is constant, it was determined that when their peaks were sharpened and their width decreased, maximum deformation, maximum stress, and maximum strain affecting the cell walls of both bacteria increased. When sharpness and width of the nano-patterns are kept constant and the spacing is increased, maximum deformation, maximum stress, and maximum strain in P. aeruginosa cell walls increase, but a decrease in S. aureus was observed. This study proves that changes in the geometric structures of nanopatterned surfaces can show different effects on different bacteria.

Predicting uniaxial compressive strength of tuff after accelerated freeze-thaw testing: Comparative analysis of regression models and artificial neural networks

Predicting uniaxial compressive strength of tuff after accelerated freeze-thaw testing: Comparative analysis of regression models and artificial neural networks

Standoff Identification of Plastic Waste Using a Low-Cost Compact Laser-Induced Breakdown Spectroscopy (LIBS) Detection System.

We report the standoff/remote identification of post-consumer plastic waste by utilizing a low-cost and compact standoff laser-induced breakdown spectroscopy (ST-LIBS) detection system. A single plano-convex lens is used for collecting the optical emissions from the plasma at a standoff distance of 6.5 m. A compact non-gated Czerny-Turner charge-coupled device (CCD) spectrometer (CT-CCD) is utilized to analyze the optical response. The single lens and CT-CCD combination not only reduces the cost of the detection system by tenfold, but also decreases the collection system size and weight compared to heavy telescopic-based intensified CCD systems. All the samples investigated in this study were collected from a local recycling plant. All the measurements were performed with only a single laser shot which enables rapid identification while probing a large number of samples in real time. Furthermore, principal component analysis has shown excellent separation among the samples and an artificial neural network analysis has revealed that plastic waste can be identified within ∼10 ms only (testing time) with accuracies up to ∼99%. Finally, these results have the potential to build a compact and low-cost ST-LIBS detection system for the rapid identification of plastic waste for real-time waste management applications.

Comparative studies of Taguchi dynamic analysis, statistical, and artificial neural networks for low-carbon steel corrosion inhibition in acidic media

The aim of the present work is to optimize the corrosion inhibition process of low-carbon steel in 2.5 M phosphoric acid at different temperatures and inhibitor concentrations. The weight loss method is used to evaluate the corrosion rate in the absence or presence of a corrosion inhibitor. Taguchi dynamic (L16), mathematical regression, and artificial neural networks (ANN) are used during the data processing. The outcomes showed that the %IE increased with inhibitor concentration and temperature up to 50 oC. The Taguchi method showed that the optimum conditions for the corrosion inhibition temperature were found to be 50 oC and a 0.05 M KI concentration. Two mathematical models are proposed to construct a relationship between %IE and independent variables: the exponent model (EM) and the polynomial model (PM). PM was more accurate than EM, with a significant correlation coefficient approaching 0.9851. Temperature had a greater effect on the %IE than inhibitor concentration, which was consistent with Taguchi's findings. In ANN analyses, five networks were suggested: linear, a generalized regression neural network (GRNN), radial basis functions (RBF), and two multi-layer perceptron (MLP) networks. The highest correlation coefficient (0.996) and lower absolute error (0.126) were achieved by MLP 2:2-9-4-1:1.

Phosphate removal from aqueous solutions using reversal mode electrocoagulation with iron and aluminum electrodes: RSM optimization and ANN modeling

Abstract Phosphate pollution significantly contributes to eutrophication and the degradation of aquatic ecosystems. The removal of phosphate from wastewater before discharging into the environment is essential for the sustainability of the ecosystem. This work focuses on using a polarity reversal mode electrocoagulation (PRM-EC) system integrated with iron (Fe) and aluminum (Al) electrodes to remove phosphate from wastewater. The conditions for the removal process were optimized using response surface methodology (RSM). Central composite design (CCD) was used to design the experiment and numerical optimization was utilized to find the optimal conditions. The phosphate removal efficiency could reach 93.12% at a current density of 40 A m−2, time of 30 min, pH of 6.4, and electrode distance of 10.5 mm. The energy consumption was about 0.4 kW m−3. The artificial neural network (ANN) modeling showed that the current density was the most influencing factor, followed by time, pH, and electrode distance. The mechanism underlying the PRM-EC process encompassed electrode dissolution, floc formation, phosphate adsorption, and precipitation. The findings in the work show that PRM-EC is an environmentally friendly and effective solution for phosphate removal.

An Experimental and ANN Analysis of Ammonia Energy Integration in Biofuel Powered Low-Temperature Combustion Engine to Enhance Cleaner Combustion

In the pursuit of global net-zero emissions targets and cleaner combustion technologies, researchers are increasingly turning to innovative solutions. Ammonia, as a carbon-free fuel, holds promise for achieving low emissions and efficient combustion. This study explores the potential synergies of various ammonia shares (20%, 40%, and 60%) combined with Citronella biofuel in a Low-Temperature Combustion (LTC) Reactivity Controlled Compression Ignition (RCCI) engine. Comparative analysis between a 40% ammonia blend Citronella biofuel (A40) and a Standard Diesel (SDL) in an RCCI engine demonstrates significant reductions in oxides of nitrogen (by 17%), carbon dioxide (by 6.6%), and smoke (by 9.8%), along with a notable enhancement in Brake Thermal Efficiency (BTE) by 5.4% for A40. The validation of results using an Artificial Neural Network (ANN) model shows robust coefficients of determination (97%), high R values (0.9945 to 0.9995), and low Root Mean Square Error values (ranging from 0.0219 to 0.0483). These findings strongly support the efficacy of the A40 blend, positioning it as a promising and viable alternative fuel for achieving cleaner combustion in LTC-RCCI engines.

Assessing the Relationships between Physical Activity, a Healthy Life, and Personal Happiness in European Union Countries.

Background: Maintaining a physically active lifestyle is a determinant factor of a healthy life and personal happiness. Meanwhile, physical inactivity remains a significant issue, resulting in negative consequences for public health. Objectives: This paper investigates the relationships between physical activity, physical inactivity, a healthy life, life expectancy, and personal happiness in European Union (EU) countries. Methods: This empirical study uses an artificial neural network and cluster analysis to analyze and interpret data from 27 EU countries. Artificial neural network analysis enables the assessment of the relationships between physical activity and inactivity, a healthy life, and personal happiness, while cluster analysis identifies groups of EU countries based on physical activity, healthy life, and personal happiness indicators. Results: The results show significant positive links between physical activity and improvements in healthy living and personal happiness. Conclusions: This study highlights considerable variations among EU countries regarding the levels of physical activity, healthy living, and personal happiness, emphasizing the importance of promoting physical activity to enhance public health and overall well-being. The findings suggest the need to develop customized policies that address country-specific factors and promote an active lifestyle.

Distinct soluble immune checkpoint profiles characterize COVID-19 severity, mortality and SARS-CoV-2 variant infections.

Over the past four years, the COVID-19 pandemic has posed serious global health challenges. The severe form of disease and death resulted from the failure of immune regulatory mechanisms, closely highlighted by the dual proinflammatory cytokine and soluble immune checkpoint (sICP) storm. Identifying the individual factors impacting on disease severity, evolution and outcome, as well as any additional interconnections, have become of high scientific interest. In this study, we evaluated a novel panel composed of ten sICPs for the predictive values of COVID-19 disease severity, mortality and Delta vs. Omicron variant infections in relation to hyperinflammatory biomarkers. The serum levels of sICPs from confirmed SARS-CoV-2 infected patients at hospital admission were determined by Luminex, and artificial neural network analysis was applied for defining the distinct patterns of molecular associations with each form of disease: mild, moderate, and severe. Notably, distinct sICP profiles characterized various stages of disease and Delta infections: while sCD40 played a central role in all defined diagrams, the differences emerged from the distribution levels of four molecules recently found and relatively less investigated (sCD30, s4-1BB, sTIM-1, sB7-H3), and their associations with various hematological and biochemical inflammatory biomarkers. The artificial neural network analysis revealed the prominent role of serum sTIM-1 and Galectin-9 levels at hospital admission in discriminating between survivors and non-survivors, as well as the role of specific anti-interleukin therapy (Tocilizumab, Anakinra) in improving survival for patients with initially high sTIM-1 levels. Furthermore, strong associations between sCD40 and Galectin-9 with suPAR defined the Omicron variant infections, while the positive match of sCD40 with sTREM-1 serum levels characterized the Delta-infected patients. Of importance, this study provides a comprehensive analysis of circulatory immune factors governing the COVID-19 pathology, and identifies key roles of sCD40, sTIM-1, and Galectin-9 in predicting mortality.

Impact of Soybean Biodiesel Blends with Mixed Graphene Nanoparticles on Compression Ignition Engine Performance and Emission: An Experimental and ANN Analysis

The extensive use of fuels in power generation plants, industries, and transportation has led to a scarcity of fossil fuels and has contributed to global warming. This has prompted researchers to focus on improving internal combustion engine performance, as the transportation system accounts for 50% of global fuel consumption. Soybean biodiesel plays a vital role in reducing emissions and fuel consumption in engines. In the current work, a blend of soybean oil is used as a biodiesel (B20, B40, and B60), with and without the addition of graphene nanoplatelets (GNPs) examined on a constant-speed compression ignition (CI) engine with respect to pure diesel. The blends are denoted as B20GNP10, B20GNP20, B20GNP40, B20GNP60, B20GNP80, and B20GNP100, according to their proposition of biodiesel and graphene nanoplatelets. Similar blends were prepared for B40 and B60 combinations with similar GNPs concentrations. The prepared blend properties were measured, and good thermophysical properties were found. The trial includes testing the engine emissions and performance at altering loads ranging from 0 to 12 kg for all blends. The artificial neural network (ANN) tool is used to forecast the accuracy of experimental results. Compared to pure diesel, the B60GNP100 blend at 12 kg load condition showed the lowest brake specific fuel consumption at 12.58 % and the highest brake thermal efficiency at 27.13%. Emissions were estimated using a gas analyzer, and the outcomes indicated that the biodiesel blends have controlled levels of CO, CO2, NOx, and HC compared to pure diesel. The ANN model with 99.99% accuracy was developed using experimental data, confirming the accuracy of the experimental results with lower simulation time and cost. Additionally, the B60GNP100 blend yielded better results compared to previous studies.

A conceptual framework for exploring the use of a green marketing mix and brand image to attract customers for green services while visiting a hotel: A step toward marketing sustainability

Recently, the application of green marketing strategies to promote marketing sustainability has increased in response to the growing demand for eco-friendly products, Price, promotion, place, and service are all variables that influence purchasing decisions, and the current study aims to confirm this. as the brand plays a significant role as well because it is regarded as the driving force behind consumer attraction in society. For this study, a framework connecting brand image (brand benefits and competence, brand identity, brand personality, brand association, brand attitude, and behavior), the green marketing mix (green services, green place, green price, green promotion), and green purchasing decisions were created to better understand the relationships among these variables. Artificial neural network (ANN) analysis and structural equation modeling (SEM) based on deep learning will be conducted via the new hybrid analysis approach. The research employed a survey of hotel customers in Iraq, Additionally, the researchers conducted structured interviews with a group of hotel managers. The issue of green marketing and how it relates to environmentally conscious consumer choices has received much attention, which has made addressing sustainability a top priority in the hospitality industry, particularly in hotels. In this study, customers were the main focus since their choice is crucial regarding switching to green services. However, a group of these hotel managers were interviewed to validate the findings and establish credibility. The current study was limited since it focused on clients in general, without considering clients from outside Iraq. Future studies can examine the variable of organizational support and the degree of change acceptance among customers.

Classification of wheat flour levels in powdered spices using visual imaging

The present research aims to investigate the ability of image processing in the detection of wheat flour adulterant in different spices. Different samples of black pepper, red pepper, and cinnamon were prepared with different levels of wheat flour adulteration. The images of the samples were captured using a visible camera. An image processing algorithm was designed and codded in MATLAB software to process the acquired images. Different color and textural features were extracted. The efficient features were selected and classified using artificial neural network and support vector machine. In the present research, the classification accuracies of artificial neural network analysis for black pepper, red pepper, and cinnamon products were 97.8, 100.00, and 100.00 %, respectively. During the experiment, these results were better than the performance of the support vector machine, (93.33, 100.00, and 98.88 %, respectively). Visible image processing can be used accurately to detect different adulterant levels that increase product quality while reducing operating costs.

Determination of electron screening potential of 6Li(p, [formula omitted])3He reaction using Multi Layer Perceptron based neural network

Understanding nuclear reactions between light-charged nuclei in the sub-Coulomb energy region is crucial for several astrophysical processes. Accurate determination of the reaction cross-section within the astrophysically important Gamow range is challenging due to electron screening. Various methods, including polynomial fits, R-Matrix, and the indirect Trojan Horse Method (THM), have estimated electron screening energies that exceed the adiabatic limit. This study aims to derive the bare astrophysical S-factor for the reaction 6Li(p,α)3He and to extract electron screening energies using Multi-Layer Perceptron-based Artificial Neural Network (ANN) analysis. Experimental S-factors for 6Li(p,α)3He, obtained from the literature, are reanalyzed with the ANN algorithm to determine the energy-dependent S-factor. The bare astrophysical S-factor is calculated from data above 60 keV, where electron screening is negligible. The electron screening potential is then derived by comparing the shielded S-factor with the bare S-factor. The ANN-based analysis yields an electron screening potential of 220 eV, suggesting that ANN could be a viable tool for estimating electron screening potentials in light nuclei reactions.

Use of machine learning methods in predicting the main components of essential oils: Laurus nobilis L.

Laurus nobilis L. is an export plant that contributes to the industry with its medicinal and aromatic properties, as well as the use of its leaves or fruits in the food, pharmaceutical, and cosmetic industries. This study aims to predict the first three main components of bay laurel essential oil with different machine learning methods as a result of GC-MS analysis, taking into account a wide range of parameters used before and after the analysis of the essential oil in pharmacognostic studies. In this study, the K-Means algorithm was used to detect unknown relationships (chemical or physical properties) between the data obtained as a result of the analysis, and Artificial Neural Network (ANN) and Decision Tree machine learning methods were used to perform prediction operations. Based on this, considering the R2 values obtained from ANN analyses, R2 values of 0.9992 for testing, 1 for training, 1 for validation value, and 0.99992 for all were obtained. The ANN model produced results close to the real values with an accuracy rate of 98.97% in predicting the three main components. The Decision Tree algorithm achieved classification with a 96.23% accuracy rate in predicting the three main components. K-Means clustering was performed with an accuracy rate of 97.83% and significant relationships between the features of the Laurus nobilis dataset were detected. This comprehensive study serves as a guide for machine learning research on other essential oils.

Non-similar solution of Casson fluid flow over a curved stretching surface with viscous dissipation; Artificial neural network analysis

PurposeThe main focus is to provide a non-similar solution for the magnetohydrodynamic (MHD) flow of Casson fluid over a curved stretching surface through the novel technique of the artificial intelligence (AI)-based Lavenberg–Marquardt scheme of an artificial neural network (ANN). The effects of joule heating, viscous dissipation and non-linear thermal radiation are discussed in relation to the thermal behavior of Casson fluid.Design/methodology/approachThe non-linear coupled boundary layer equations are transformed into a non-linear dimensionless Partial Differential Equation (PDE) by using a non-similar transformation. The local non-similar technique is utilized to truncate the non-similar dimensionless system up to 2nd order, which is treated as coupled ordinary differential equations (ODEs). The coupled system of ODEs is solved numerically via bvp4c. The data sets are constructed numerically and then implemented by the ANN.FindingsThe results indicate that the non-linear radiation parameter increases the fluid temperature. The Casson parameter reduces the fluid velocity as well as the temperature. The mean squared error (MSE), regression plot, error histogram, error analysis of skin friction, and local Nusselt number are presented. Furthermore, the regression values of skin friction and local Nusselt number are obtained as 0.99993 and 0.99997, respectively. The ANN predicted values of skin friction and the local Nusselt number show stability and convergence with high accuracy.Originality/valueAI-based ANNs have not been applied to non-similar solutions of curved stretching surfaces with Casson fluid model, with viscous dissipation. Moreover, the authors of this study employed Levenberg–Marquardt supervised learning to investigate the non-similar solution of the MHD Casson fluid model over a curved stretching surface with non-linear thermal radiation and joule heating. The governing boundary layer equations are transformed into a non-linear, dimensionless PDE by using a non-similar transformation. The local non-similar technique is utilized to truncate the non-similar dimensionless system up to 2nd order, which is treated as coupled ODEs. The coupled system of ODEs is solved numerically via bvp4c. The data sets are constructed numerically and then implemented by the ANN.

Comparison of machine learning algorithms and multiple linear regression for live weight estimation of Akkaraman lambs

This study was designed to predict the post-weaning weights of Akkaraman lambs reared on different farms using multiple linear regression and machine learning algorithms. The effect of factors the age of the dam, gender, type of lambing, enterprise, type of flock, birth weight, and weaning weight was analyzed. The data was collected from a total of 25,316 Akkaraman lambs raised at multiple farms in the Çiftlik District of Niğde province. Comparative analysis was conducted by using multiple linear regression, Random Forest, Support Vector Machines (and Support Vector Regression), Extreme Gradient Boosting (XGBoost) (and Gradient Boosting), Bayesian Regularized Neural Network, Radial Basis Function Neural Network, Classification and Regression Trees, Exhaustive Chi-squared Automatic Interaction Detection (and Chi-squared Automatic Interaction Detection), and Multivariate Adaptive Regression Splines algorithms. In this study, the test dataset was divided into five layers using the K-fold cross-validation method. The performance of models was compared using performance criteria such as Adjusted R-squared (Adj-\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${R}^{2}$$\\end{document}), Root Mean Square Error (RMSE), Mean Absolute Deviation (MAD), and Mean Absolute Percentage Error (MAPE) by utilizing test populations in the predicted models. Additionally, the presence of low standard deviations for these criteria indicates the absence of an overfitting problem. \\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${R}^{2}$$\\end{document}The comparison results showed the Random Forest algorithm had the best predictive performance compared to other algorithms with Adj-\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${R}^{2}$$\\end{document}, RMSE, MAD, and MAPE values of 0.75, 3.683, 2.876, and 10.112, respectively. In conclusion, the results obtained through Multiple Linear Regression for the live weights of Akkaraman lambs were less accurate than the results obtained through artificial neural network analysis.

Determinants and Outcomes of Green Technology Innovation Adoption among Third-Party Logistics Firms in China: A SEM-ANN Analysis

Although the adoption of green innovation has been explored in various contexts, there is still a lack of research on the determinants of green technology innovation adoption (GTIA) in the third-party logistics (3PL) industry of emerging economies and the resulting economic, environmental, operational, and intangible outcomes. This study aims to investigate the influence of different determinants of technological, organizational, and environmental factors in the TOE-DOI framework on the adoption of green technology innovation, as well as the outcomes of such adoption. The study utilizes sample data from 544 Chinese 3PL firms and employs two-stage structural equation modeling and artificial neural network analysis. Partial least squares structural equation modeling (PLS-SEM) explains 85.4% of the variance in green technology innovation adoption, 21% in environmental outcomes, 18.5% in economic outcomes, 20.8% in operational outcomes, and 22.3% in intangible outcomes. The artificial neural network (ANN) model ranks the standardized importance of each predictive variable. The results indicate that institutional pressure is the most significant determinant of GTIA. Additionally, 3PL firms should consider the positive impact of green supplier integration and relative advantage. Complexity does not have a positive impact on GTIA. The longer a company has been established, the more experience and resources it accumulates, and the more opportunities it has to adopt green technology innovation. This study contributes to the existing research on emerging economies and other regions. Furthermore, this is the first study to successfully validate the nonlinear relationship within the Technology-Organization-Environment (TOE) framework and diffusion of innovation (DOI) theory, namely the TOE-DOI framework. The research findings further enhance the current understanding of green technology innovation adoption and its impact. This study provides valuable insights for managers and policymakers in the 3PL industry to achieve various effects, such as environmental, economic, operational, and intangible outcomes.

A new strategy for manufacturing, modeling, and optimization of 3D printed polylactide based on multiple nonlinear neuro regression analysis and stochastic optimization methods

Although there are many studies related to design, modeling, and optimization of fused deposition modeling (FDM) process parameters in the literature, the absence of a systematic approach to increase the reliability of model selection and optimization results is an important shortcoming that must be addressed. To make up for this deficiency, a new strategy was proposed to obtain desired quality on mechanical properties by adjusting FDM process parameters. This attempt involves manufacturing, modeling, and optimization point of view. The D-optimal method was employed to form experiment set, including process parameters. A hybrid approach neuro regression combining artificial neural network (ANN) and regression analysis together was used for modeling of FDM process. The most significant advantage of the neuro-regression approach compared to ANN is that the mathematical models can be used directly without needing any transformation. This is not possible in neural networks and, therefore, significantly limits and complicates the use of models obtained using ANN. The present study aims at optimization of the FDM process parameters, including infill density, infill pattern, layer thickness, and print speed on ultimate strength, fracture strength, and fracture strain for polylactide (PLA). In this regard, modified versions of the optimization algorithms Differential Evolution, Nelder Mead, and Simulated Annealing were used to find the best or elite designs. Linear or nonlinear models consisting of polynomial, trigonometric, and logarithmic expressions and their hybrid forms were employed to define the strength and strain behavior of PLA. It is concluded that (a) implementations of the optimization algorithms provide a 19% improving the minimum strain value if it is compared with the experimental results, (b) infill pattern types ( x2) were found as honeycomb, triangle, and cubic for the designs in terms of maximum fracture strength, minimum strain, and maximum ultimate tensile strength respectively, (c) many alternatives near optimum local designs could be obtained based on Nelder Mead algorithm for fracture strength and ultimate strength parameters. Thus, this allows work in a wide range of applications without depending on a single result for production.

Exploring cognitive-behavioral drivers impacting consumer continuance intention of fitness apps using a hybrid approach of text mining, SEM, and ANN

Amidst the digital revolution, consumer wellness services have been transformed by the proliferation of health and fitness apps (HFAs). Leveraging the information systems continuance theory, this research investigates three primary cognitive and behavioral drivers involving satisfaction, attitude, and habit that impact continuance intention. It also examines the underlying antecedents of these drivers. The investigation into what shapes users' attitudes incorporates elements from the health belief model and the dual passion theory. Additionally, to determine what drives user satisfaction, the study analyzes online consumer reviews through structural topic modeling. This study utilizes a combination of topic modeling, PLS-SEM, and Artificial Neural Network (ANN) analysis for development and validation of conceptual model. After conducting a survey and analyzing 367 responses via PLS-SEM, the findings indicated that harmonious passion, health consciousness, and perceived disease threat greatly influenced users' attitudes toward these apps. The results emphasized satisfaction, attitude, and habit formation as crucial determinants of continued app use. An additional key contribution to the literature was the exploration of habit as a partial mediator between attitude and satisfaction in determining continued app usage, which has remained unexplored in the HFAs context. A comparison of results between the PLS-SEM and ANN analysis showed that there is a mismatch in the rankings of security and customization on satisfaction. This implies that there are possibly hidden factors that could affect security and customization differently in real-world settings. The study concludes by offering practical recommendations for managers of HFAs to enhance consumer service experiences and drive continuance intention.