ABSTRACT Inverters are essential for reducing total harmonic distortion (THD) and power losses, while improving system efficiency. In this study, a three‐level neutral point clamped (NPC) inverter using space vector pulse width modulation (SVPWM) is presented. The primary objective is to perform regression‐based sector prediction for a three‐level NPC inverter. Sector prediction performance is evaluated using linear, exponential, and polynomial regression methods. The linear regression achieved the best performance with 0.9722 R 2 , 0.2452 RMSE, and 0.2846 MAE, outperforming the polynomial (0.9586 R 2 ) and exponential (0.8712 R 2 ) models. Quantitatively, the linear model provides a 14.4% accuracy improvement over the polynomial model and a 51.5% improvement over the exponential model. Furthermore, the inverter output shows low distortion, achieving 0.29% THD in simulation and 1.9% THD in experimental measurements. The model is analyzed for output current, voltage, and THD. The results show a high consistency between simulation, experimental validation, and regression models. This approach reduces the computational complexity of three‐level NPC inverters. The results confirm both the theoretical accuracy and practical applicability of the proposed model and illustrate its potential to improve power conversion efficiency and power quality.

Abstract Background With the rapid changes in lifestyle and pace, the public is facing multiple pressures in social and work planning, which not only affect their mental health but may also have a significant impact on their dietary behavior. The relationship between stress and dietary behavior may be influenced by mediating variables such as emotional state and sleep quality, but existing research on these mechanisms is insufficient and lacks consistent conclusions. Therefore, the systematic exploration of the correlation between stress perception and dietary behavior in psychological disorders aims to reveal the chain mediated role of anxiety and sleep quality between stress perception and dietary behavior, and provide scientific basis and decision-making reference for dietary behavior intervention measures in psychotherapy activities. Methods The study used convenience sampling to select undergraduate students who had participated in stress management in universities as the research subjects. A questionnaire survey was conducted using the Perceived Stress Scale (PSS), Self Rating Anxiety Scale (SAS), and Dietary Behavior Scale. After obtaining the survey results, one-way ANOVA and multiple linear regression methods were used to explore the relationship between stress, anxiety, and dietary behavior. At the same time, the study used analysis of variance (ANOVA) to test whether the differences in the scores of dietary behavior and healthy eating awareness among students with different exercise frequencies were statistically significant. Results The experimental results showed a positive correlation between stress perception and eating behaviors such as snacking (0.17, p<.01), food response (0.16, p<.01), emotional eating (0.327, p<.01), restrictive eating (0.099, p<.01), and picky eating (0.222, p<.01). There is a negative correlation between perceived stress and awareness of healthy eating (-0.216, p<.01). There is a positive correlation between anxiety and eating behaviors such as snacking (0.181, p<.01), food response (0.135, p<.01), emotional eating (0.292, p<.01), restrictive eating (0.082, p<.01), and picky eating (0.197, p<.01). There is a negative correlation between anxiety and awareness of healthy eating (-0.139, p< .01). At the same time, college students who exercise more than 3 times a week scored lower in picky eating and emotional eating compared to the group of students who exercise very little, while scoring higher in restrictive eating and awareness of healthy eating compared to the group of students who exercise very little. Experiments have shown that when the subjects' perceived stress or anxiety levels increase, unhealthy eating behaviors also tend to increase. When an individual's awareness of healthy eating increases, their stress perception and anxiety levels may decrease. Discussion The research findings indicate a significant bidirectional relationship between stress, anxiety, and dietary behavior. Stress and anxiety not only affect dietary behavior, but changes in dietary behavior can also in turn affect an individual's levels of stress and anxiety. This discovery provides a new perspective for the prevention and treatment of psychological disorders, emphasizing the importance of considering dietary behavior when treating psychological disorders. The research also provides a theoretical basis for developing comprehensive psychological intervention measures, which may include stress management, anxiety treatment, and dietary behavior adjustment. Future research can further explore how to alleviate stress and anxiety by improving dietary behavior, as well as how to improve dietary behavior through stress and anxiety management.

This research aims to see the effect of Total Asset Turnover and Debt To Equity Ratio (DER) on Return On Assets (ROA) at PT Indofood Sukses Makmur Tbk which is listed on the BEI in 2010 - 2023. The data used in this research is data secondary in the form of the annual financial report of the company under study. This research uses quantitative data and the data source used is secondary data with an analysis method using multiple linear regression methods with data processing using SPSS v.25. The results of hypothesis testing (T Test) partially state that the Total Asset Turnover variable does not have a significant and positive influence on Return On Assets (ROA) and the Debt To Equity Ratio (DER) variable has a significant and negative influence on Return On Assets (ROA). Simultaneous F Test results of the Total Asset Turnover and Debt To Equity Ratio (DER) variables show a significant influence between Total Asset Turnover and Debt To Equity Ratio (DER) on Return On Assets (ROA).

Corrosion of steel structures in marine environments is a critical issue affecting infrastructure integrity and maintenance costs worldwide. Generally, inhibitors have proven to reduce the corrosion rate to the barest minimum than other methods. The inhibitors are produced using the experimental method which is time consuming and costly. This necessitate the development of models for the quick assessment of the efficiency of the inhibitor. This research focused on the prediction of corrosion inhibitory efficiency of water hyacinth on mild steel in marine environment using multiple linear regression (MLR) method. Various concentrations (5 ml, 10 ml, 15 ml, 20 ml and 25 ml) were added to the samples immersed in seawater and a sample without the addition of the inhibitor was used as the control for a period of 30 days. The study was carried out using weight loss method and the corrosion rate as well as the inhibition efficiency were calculated. Phytochemical analysis and atomic absorption spectroscopic were carried out on the inhibitor while Scanning Electron Microscopy and Energy Display X-ray Spectroscopy were used to analyze the steel sample. The analysis of the result showed that the best inhibition efficiency obtained was 90% and this was achieved with 15% concentration of the inhibitor. Multiple linear regression model was developed to predict the inhibitor’s efficiency. The predicted efficiency with the MLR model was compared with that of the experimentally obtained efficiency and the outcome shows a conformity between the experimental and the predicted value. It would therefore be recommended to rely on multiple linear regression in predicting the efficiency of water hyacinth for corrosion control of mild steel in marine environment based on the closeness of the predicted values to the experimental values.

This paper proposes an integrated decision analysis framework combining the Analytic Hierarchy Process (AHP) with a grey prediction model, focusing on a unified modeling approach for matching different entities and forecasting trends under multi-objective constraints. First, a multi-level structural model based on AHP is constructed. Starting from weight allocation and consistency testing, it quantitatively evaluates the comprehensive performance of different entities under multi-dimensional indicators, enabling systematic identification of optimal matching targets. Second, a grey prediction model is introduced to model and forecast the trend changes of key indicators under limited sample conditions. The model's reliability and applicability are validated through accuracy testing and residual analysis, thereby delineating the system's future evolutionary direction. Finally, combining linear regression and time series analysis methods, the model quantitatively evaluates the effects of implemented intervention measures, forming a complete analytical process from prediction to decision-making to effect verification. This model features a clear structure, stable computation, and low sample size requirements. It enables effective prediction and decision support under conditions of incomplete information, demonstrating strong universality and promotional value. It provides a feasible and efficient modeling approach for the comprehensive analysis of similar complex systems.

Stock market predictions, being one of the most important areas in financial studies, have for long been sought after by many researchers in these areas, with their applications being of utmost importance in studies on theory. Here, in an endeavor to bring about an efficient yet interpretative model for stock market predictions, there has been an impressive amalgamation of approaches based on linear regression models, matrix theory, and error correction mechanisms. Here, in particular, while model parameters have been identified with utmost precision based on least squares, there have been improvements with regard to computational speeds based on matrix calculation, with improved market versatility for resistance to dynamic markets based on error correction mechanisms. On testing applications, there has been an analysis based on historical Google stock market performances. It has been determined in these studies, based on their applications, that there has indeed been remarkable stability in performance levels in identifying stock performance based on its predictive nature, while particularly emphasizing their high flexibility levels in real-time applications based on its localization levels in its sliding window-based method. Though there have been limitations with regard to response levels based on short-term market changes, there have indeed been many advantageous levels in computational simplicity, interpretability, and versatility based on its applications in predictive conclusions for financial dealers. These studies have indeed paved new ways in achieving efficient fusion in financial predictive studies, with appropriate applications in their development in computational finance based on innovations brought about in finance theory based on modern computing innovations.

This study aims to analyze the influence of product quality, brand image, and promotion on purchasing decisions for Honda BeAT motorcycles in Denpasar City. This study uses a quantitative approach with a causality research design to test the causal relationship between variables. The sampling technique used was purposive sampling with a total of 150 respondents, namely Honda BeAT motorcycle users who have income and are domiciled in Denpasar City. Data collection was carried out through the distribution of structured questionnaires compiled using a five-point Likert scale to measure respondents' perceptions of the research variables. The data obtained were then analyzed using multiple linear regression methods to determine the influence of each independent variable on purchasing decisions, both partially and simultaneously. The results showed that product quality, brand image, and promotion had a positive and significant influence on purchasing decisions for Honda BeAT motorcycles. Partially, each variable has a real contribution in encouraging consumers to make purchases. Simultaneously, the three variables together have a significant influence on purchasing decisions. These findings indicate that improving product quality, strengthening positive brand image, and implementing effective promotional strategies are important factors in increasing consumer interest and purchasing decisions for the Honda BeAT in Denpasar City.

Tyre wear is a core factor affecting driving safety and operating costs. Conventional studies have focused on traditional factors such as tyre materials, suspension systems and driving behavior, while neglecting the deeper mechanisms by which vehicle exterior design affects tyre wear through aerodynamic pathways. This study aims to construct a theoretical framework to reveal how aerodynamic design parameters (e.g., front and rear wings, diffusers, etc.) affect tyre friction and slip rate by changing the downforce distribution, and ultimately affect tyre wear rate and service life. By integrating Bernoulli's equation and the multiple linear regression method, a tyre life prediction model covering key variables such as friction, average speed, underbody height, road condition and slip rate is established, and the differential wear mechanisms such as thermo-mechanical fatigue and grinding loss caused by front and rear axle underpressure imbalance are also deeply analyzed. The results show that the aerodynamic effect induced by exterior design has a decisive influence on tyre wear patterns, and the theoretical model proposed in this paper provides an innovative analytical framework for quantitatively assessing this effect. Although the specific coefficients of the model are yet to be calibrated and verified by CFD simulation and real vehicle test data, the theoretical results of this study are of great significance for optimizing vehicle design, extending tyre life and reducing operating costs.

Accurate characterization of river channel geometry is essential for hydrological and hydraulic analyses, yet the increasing use of unmanned aerial vehicle (UAV) photogrammetry introduces challenges related to uneven point density, shadow-induced data gaps, and spurious outliers. This study proposed a novel approach for reconstructing 3D river channels from UAV-derived point clouds, emphasizing K-nearest neighbor local regression (KLR), and compared it with the LOWESS model. Method performance was examined through controlled simulations of trapezoidal, triangular, and U-shaped synthetic channels, where KLR consistently preserved morphological fidelity and produced lower RMSE than LOWESS, particularly at channel bends and bed undulations, while a neighborhood selection heuristic approach demonstrated robust results across varying data densities. Synthetic channel experiments show that the proposed K-nearest-neighbor local linear regression (KLR) method achieves RMSE values below 0.06 all tested geometries. In contrast, LOWESS produces substantially larger errors, with RMSE values exceeding 0.9 across all channel shapes. Subsequent application to two South Korean field sites reinforced these findings. In the data-scarce Migok-cheon stream, KLR effectively interpolated missing surfaces while maintaining geomorphic realism, whereas LOWESS generated over-smoothed representations. Within the dense Ogsan Bridge dataset, KLR retained small-scale bed features critical for hydraulic simulations and cross-sectional delineation, while LOWESS obscured local variability. Conclusively, the results demonstrate that KLR provides a more reliable and computationally efficient framework for UAV-based 3D river channel reconstruction, with clear implications for hydraulic modeling, flood risk management, and the advancement of digital-twin systems in operational hydrology.

This study aims to empirically examine and analyze the influence of Brand Image and Service Quality on Customer Satisfaction at PT Sejahtera Surya Intramedika within the context of a Business-to-Business (B2B) pharmaceutical distributor. This research is crucial given the intensifying competition in the healthcare industry and the mandatory compliance with strict drug distribution regulations. Employing a quantitative causality approach, this study collected data from 101 customer respondents (pharmacies) determined through a multistage sampling technique combining cluster and purposive sampling based on operational area criteria including Blitar City, Kediri, Tulungagung, and Pare, as well as a specific minimum transaction volume. Data analysis was conducted using multiple linear regression methods assisted by SPSS software to test the hypotheses. Statistical test results indicate that partially, Brand Image has a positive and significant effect on customer satisfaction with a regression coefficient of 0.347. Service Quality was also proven to have a positive and significant effect, and was identified as the most dominant variable with a coefficient of 0.620. Simultaneously, both variables contributed 69.6% influence to the variation in customer satisfaction, as validated by the calculated F-value of 112.166. This study concludes that the synergy between a professional brand image and responsive service quality is a fundamental key to company success. Practical implications suggest that management should prioritize investment in logistics reliability and personnel competence to strengthen strategic partnerships and ensure sustainable customer satisfaction amidst the dynamics of the competitive pharmaceutical market.

ABSTRACT The mountain chain of the Western Ghats features geomorphic characteristics of immense importance, with unique biophysical and ecological processes and exceptionally high levels of biological diversity and endemism. This study conducts a phytosociological analysis and estimates carbon and biomass storage using linear regression methods for various types of vegetation. The r2 value remains different for each vegetation category. It also evaluates the ecological importance of the Velliyamattam – Kanjar green isle (known as ‘Pachathurut’ in the Malayalam language) through estimates of its carbon and biomass. The biomass of the studied plots was predicted using an equation generated from the observed biomass and the average NDVI (derived from Landsat 8 OLI imagery). Biomass was 5.36 t 0.09 ha−1 in a 13-year-old plantation and 4.17 t 0.09 ha−1 in a 3-year-old Pachathurut plot. The findings indicate that Pachathurut shows a higher carbon stock within a short period than conventional plantations.

The development of rail transportation in Indonesia continues to change over time. These changes are influenced by various factors, such as government policies, the economic situation, and improvements in railway infrastructure. This dynamic suggests that better transportation planning requires predictive techniques that can accurately identify changing patterns. This study aims to compare Linear Regression and Artificial Neural Network (ANN) methods in predicting national rail passenger numbers. Before being used for modeling, the time series data underwent a preprocessing stage. The research process included dividing the data into training and test data, applying both prediction methods, and evaluating model performance using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE). The results showed that the ANN method was more accurate than the Linear Regression method. Therefore, the ANN method may be a better choice to assist rail transportation planning in Indonesia.

Dynamic chest radiography (DCR) is a recently developed low-dose pulmonary functional imaging method that can be performed in a general X-ray room. DCR provides sequential images during respiration, and the measured changes in lung area are a promising diagnostic indicator of lung function. To investigate lung volume estimation using deep learning from DCR images during respiration and evaluate its accuracy in comparison with previously proposed estimation methods. Two convolutional neural networks (CNNs), VGG19 and DenseNet121, were trained using DCR image datasets from 257 patients, with reference lung volumes derived from corresponding computed tomography (CT) images. The performance of the models was evaluated using mean absolute error (MAE) and mean absolute percentage error (MAPE), and compared against that of a conventional linear regression model. Correlation between the estimated and reference lung volumes was assessed using Pearson's correlation coefficient (r) and the degrees-of-freedom-adjusted coefficient of determination (Rf2). Forced vital capacity (FVC) was also estimated by subtracting the lung volume at maximum exhalation from that at maximum inhalation. The VGG19 and DenseNet121 models demonstrated superior performance in estimating whole lung volume (combined right and left lung) compared to the linear regression method. Specifically, MAE was 373/376mL, MAPE was 8.1%/7.9%, r was 0.88/0.90, and Rf2 was 0.76/0.80 for VGG19/DenseNet121, respectively. In contrast, the linear regression model yielded an MAE of 568mL, MAPE of 12.4%, r of 0.84, and Rf2 of 0.69. Although the Rf2 values for DCR-derived FVC using VGG19 and DenseNet121 indicated moderate correlation, the MAE and MAPE were relatively high at 1.3/1.4L and 41.1%/47.0%, respectively. The proposed deep learning-based approach for lung volume estimation from DCR images outperformed the conventional linear regression method. Further improvements in CNN model architecture and the incorporation of guided forced respiratory maneuvers may enhance the potential for image-based pulmonary function testing.

Depression is among the most prevalent mental health issues in the world. Traditional diagnostic methods rely on clinical interviews and scoring scales, which are subjective and biased. In this paper, the adopted model is based on a massive dataset with more than 27,901 student samples to conduct experiments, where this study compares three machine learning algorithms, including Random Forest (RF), Support Vector Machine (SVM) and K-Nearest Neighbor (K-NN). The methods on SVM, RF, and K-NN preprocessed, i. e., excluding outliers through the Inter Quartile Range (IQR) method, feature encoding, and Z-score normalization procedure. The experimental results shows that all models achieved good performance, with SVM having the highest accuracy followed by RF then K-NN. This study also analyzed the confusion matrix in detail. It indicates that SVM performs well in positive class recognition, while random forest excels in negative class recognition. The research results indicate that machine learning models can capture the nonlinear relationships in depression data. They can also handle complex interactions between different factors. This is something traditional linear regression methods cannot do well. This study offers important new insight for automatic depression screening. It also contributes to the development of objective diagnostic tools in mental health care.

Rapid developments in the Artificial Intelligence (AI) industry have triggered an increased need for workers with specialized competencies, which has implications for significant variations in salary levels. This research aims to analyze the factors that influence salaries in the AI sector using the multiple linear regression method. The dataset used includes 15,000 AI job vacancies with variables including job and company characteristics. The data was engineered via the one-hot encoding method and divided into two parts: training data (80%) and test data (20%). The analysis results show that the regression model is able to explain 85% of the variation in salary, with an R² value of 0.85 and a Root Mean Square Error (RMSE) of USD 23,221. The three main factors identified as having a significant influence on salaries in the AI field are work experience, company location, and the industry in which the company operates. The experience factor reflects the skills and knowledge developed over many years, which can increase productivity (Rony et al., 2023). Company location also plays an important role, as the cost of living and demand for skilled labor varies by region (Badran, 2019). Additionally, the specific industry in which an employee works influences salary, given that more developed industries can often offer higher compensation (Huang, 2025). This research makes a significant empirical contribution to the understanding of compensation structures in the AI labor market.

From a global perspective, accounting red flags play a pivotal role in preventing corporate failure by highlighting areas of financial risk that could threaten business sustainability. The paper examines the impact of accounting red flags on corporate failure in listed manufacturing firms in Nigeria with a focus on liquidity and debt capital red flags. The Z-Score used is Altman Z-Score which is the propensity of corporate failure and the salient predictors of corporate failure are aimed to be identified. The quantitative methodology was taken and it involved analysis of financial statements of a sample of 516 listed manufacturing companies. Strong linear regression methods were used to deal with heteroscedasticity and the models were stratified based on Altman Z-Score to the distress, grey, and safe regions so that strong models were used. The main results reveal that liquidity red flags have a considerable impact only on the distress area, but it has a negative impact on the safe area, hence, suggesting efficient management in less volatile companies. A red flag of debt capital reveals the importance of effective management of debt capital as the negative impact is high, and it is important to avoid the risk of failure. One of the key recommendations is that companies need to focus on the improvement of profitability rates and sound management of debt to enhance financial stability. Besides, quality auditing and strong liquidity control is unavoidable, especially to companies that are already distressed. These findings provide useful information to managers, investors, and policymakers who aim to alleviate risks of corporate failure within the Nigerian manufacturing industry.

This study aims to analyze the relationship between the number of This study aims to analyze the relationship between the number of workplace accidents and the level of membership in the Indonesian Social Security Agency for Employment (BPJS Ketenagakerjaan) in 2024. The background of this study is based on the phenomenon of an increase in the number of BPJS Ketenagakerjaan participants accompanied by an increase in the number of workplace accident reports. This condition raises critical questions about whether there is a quantitative relationship between the two variables. Using a quantitative approach and simple linear regression method, this study analyzes aggregate data per province obtained from BPJS Ketenagakerjaan and the Ministry of Manpower. The results of the analysis show a positive and significant relationship between BPJS Ketenagakerjaan membership rates and the number of recorded workplace accidents. These findings show that as the number of workers participating in BPJS Ketenagakerjaan increases, the reporting of work accidents has also become more structured and well documented. In other words, the increase in the number of reported work accidents not only reflects an increase in incidents, but also reflects an increasingly effective reporting system. This shows that the existence and coverage of BPJS Ketenagakerjaan play an important role in supporting transparency, accountability, and the formation of a better work safety culture. This role also strengthens the social protection system for workers in Indonesia, while promoting collective awareness of the importance of a safe and healthy work environment.

This study evaluates the performance of several machine learning models in predicting dissolved oxygen concentration in the surface layer of the Mar Menor coastal lagoon. In recent years, this ecosystem has suffered a continuous process of eutrophication and episodes of hypoxia, mainly due to continuous influx of nutrients from agricultural activities, causing severe water quality deterioration and mortality of local flora and fauna. In this context, monitoring the ecological status of the Mar Menor and its watershed is essential to understand the environmental dynamics that trigger these dystrophic crises. Using field data, this study evaluates the performance of eight predictive modelling approaches, encompassing regularised linear regression methods (Ridge, Lasso, and Elastic Net), instance-based learning (k-nearest neighbours, KNN), kernel-based regression (support vector regression with a radial basis function kernel, SVR-RBF), and tree-based ensemble techniques (Random Forest, Regularised Random Forest, and XGBoost), under multiple experimental settings involving spatial variability and varying time lags applied to physicochemical and meteorological predictors. The results showed that incorporating time lags of approximately two weeks in physicochemical variables markedly improves the models’ ability to generalise to new data. Tree-based regression models achieved the best overall performance, with eXtreme Gradient Boosting providing the highest evaluation metrics. Finally, analysing predictions by sampling point reveals spatial patterns, underscoring the influence of local conditions on prediction quality and the need to consider both spatial structure and temporal inertia when modelling complex coastal lagoon systems.

This study aims to analyze the effect of the General Allocation Fund (DAU), the Special Allocation Fund (DAK), and the Revenue Sharing Fund (DBH) on the Human Development Index (HDI) in Surabaya City during the period 2010–2024. A quantitative approach was employed using a multiple linear regression method based on secondary time-series data obtained from official government institutions. The results indicate that DAU, DAK, and DBH simultaneously have a significant effect on the Human Development Index. Partially, the Special Allocation Fund exhibits the strongest and most significant influence, reflecting its targeted allocation toward basic public service sectors, particularly education and health. The General Allocation Fund also has a positive and significant effect by strengthening overall regional fiscal capacity to support public service provision. In contrast, the Revenue Sharing Fund shows a positive but statistically insignificant effect on the Human Development Index. These findings suggest that improving human development outcomes in Surabaya City requires more targeted, effective, and equitable management of intergovernmental transfer funds, with a stronger emphasis on enhancing the quality of basic public services.

This article focuses on 326 small and medium-sized B2B enterprise procurement managers and supply chain managers in the Yangtze River Delta and Pearl River Delta regions, covering areas such as machinery manufacturing and industrial product wholesale. The research focuses on the pain points of digital transformation in small and medium-sized enterprise procurement. Through questionnaire surveys and multiple linear regression methods, it explores the influencing factors of artificial intelligence (AI) adoption in the procurement process (such as supplier screening and demand forecasting) and its role in procurement decision-making efficiency and cost control. The results show that perceived usefulness and supplier AI capability positively drive adoption (β=0.32, 0.28, p<0.01), Insufficient organizational resources significantly inhibits adoption (β=-0.19, p<0.05); The adoption of AI can shorten the procurement decision-making cycle by 21.3% and reduce procurement costs by 15.7%. The data is sourced from field research conducted in 2024 and public reports from the Ministry of Industry and Information Technology and the National Bureau of Statistics, providing empirical reference for B2B enterprises to optimize AI procurement strategies, solve transformation difficulties, and improve supply chain management levels.