Largest Coefficient Of Determination Research Articles

Accelerated design and fabrication of thermal protection coating via high-throughput experiments and machine learning

For the engineering applications of thermal protection materials (TPMs), the demand targets are often multiple such as high-temperature oxidation, hot corrosion, high temperature/speed ablation, etc., which is a challenging problem for developing a tailored property by an efficient and economical method. In this work, using modified silicide-based coating as the model materials, a hybrid model for the design of TPMS with tailored properties is developed through the high-throughput experiments combined with the machine learning (ML) method. Among the four data sets, XGboost model has the smallest bias in the predicted values, the largest coefficient of determination (R2) value, and the smallest mean square error (MSE). Our work exemplifies the potential of high-throughput experiments and ML-assisted design in advancing the discovery and development of novel TPMS.

ROBUST GEOGRAPHICALLY WEIGHTED REGRESSION WITH LEAST ABSOLUTE DEVIATION (LAD) ESTIMATION AND M-ESTIMATION ON GRDP OF WEST JAVA PROVINCE

Geographically Weighted Regression (GWR) is an analytical method for data that contains spatial heterogeneity effects. However, parameter estimation in the GWR model has a weakness, namely it is prone to outliers and can cause the parameter estimation to be biased. This can be overcome by the Robust Geographically Weighted Regression (RGWR) method which is more robust against the presence of outliers. This method is suitable for Gross Regional Domestic Product (GRDP) data in West Java Province, which contains outliers and also has spatial effects. The data used in this study are secondary data obtained from the Central Statistics Agency (BPS) of West Java Province. The purpose of this study is to compare the Robust Geographically Weighted Regression (RGWR) method with the Least Absolute Deviation (LAD) Estimation and M-estimation and also to find out the factors that affect the Gross Regional Domestic Product (GRDP) in West Java Province in 2021 based on the model resulting from. Selection of the best model is seen based on the value of the coefficient of determination (R2) and Mean Squared of Error (MSE). The research results show that the Robust Geographically Weighted Regression (RGWR) method with M-estimation is much more effective in estimating the distribution of GRDP in West Java Province in 2021, seen from the larger coefficient of determination and the smaller Mean Square Error (MSE). The variables that have a significant influence on GRDP in West Java Province in 2021 are the variables of foreign investment and local income.

Intramolecular boron-locking strategy induced remarkable first hyperpolarizability: role of torsion angles between donor and acceptor units.

In conventional strategies to design donor-acceptor (D-A) organic molecules with a large electronic contribution to the first hyperpolarizability (β), the effects of the torsion angles (θ1 and θ2) between donor and acceptor moieties are barely considered. To address this issue, in this work, a promising and novel intramolecular boron-locking strategy combined with the different locking groups of different acceptors to control θ1 and θ2, has been proposed to make D-A organic molecules with large β values. Intriguingly, reducing the torsion angles will make the β value of the pyridiny thiophene triphenylamine unit (Py-Th-TPA) dramatically increase up to 94%, which is mainly ascribed to the smaller θ1 and θ2 leading to lower excited energy of the crucial excited state, and enhanced charge transfer (CT) from TPA to Py-Th moieties, and finally greatly increase the donor and acceptor part contributions to β. Correlation between the difference, |θ1 - θ2| and β, provides a large coefficient of determination, R2 = 0.78, which demonstrates that |θ1 - θ2| can be regarded as a potential descriptor for designing nonlinear optics (NLO) materials with D-A architecture. Clearly, we uncovered that θ1 and θ2 play a crucial role in the performance of NLO materials with D-A fragments.

Prediction of Dexterous Finger Forces with Forearm Rotation using Motoneuron Discharges.

Motor unit (MU) discharge information obtained via electromyogram (EMG) decomposition can be used to decode dexterous multi-finger movement intention for neural-machine interfaces (NMI). However, the variation of the motor unit action potential (MUAP) shape resulted from forearm rotation leads to the decreased performance of EMG decomposition, especially under the real-time condition and then the degradation of motion decoding accuracy. The object of this study was to develop a method to realize the accurate extraction of MU discharge information across forearm pronated/supinated positions in the real-time condition for dexterous multi-finger force prediction. The FastICA-based EMG decomposition technique was used and the proposed method obtained multiple separation vectors for each MU at different forearm positions in the initialization phase. Under the real-time condition, the MU discharge information was extracted adaptively using the separation vector extracted at the nearest forearm position. As comparison, the previous method that utilized a single constant separation vector to extract MU discharges across forearm positions and the conventional method that utilized the EMG amplitude information were also performed. The results showed that the proposed method obtained a significantly better performance compared with the other two methods, manifested in a larger coefficient of determination (R2) and a smaller root mean squared error (RMSE) between the predicted and recorded force. Our results demonstrated the feasibility and the effectiveness of the proposed method to extract MU discharge information during forearm rotation for dexterous force prediction under the real-time conditions. Further development of the proposed method could potentially promote the application of the EMG decomposition technique for continuous dexterous motion decoding in a realistic NMI application scenario.

The effect of performing various technical and tactical actions on the final results of the competitive activity of highly qualified men's and women's teams in 3x3 basketball

The article examines the peculiarities of the implementation of technical and tactical actions by highly qualified men's and women's teams at international competitions. The priority influence of the implementation of certain technical and tactical actions by highly qualified teams on the success of their performances both in individual matches and in the tournament as a whole was determined. The final success of the men's teams in the competition depends on the efficiency of long-range shots (r= -0.520), scored (r= -0.772) and missed (r= 0.635) points in the match, as well as turnovers (r= 0.502).
 For women's teams, the greatest influence on the final competitive result was scored (r= -0.678) and missed (r= 0.664) points in the match, the realization of one-point shots (r= -0.517), defensive rebounds (r= -0.305) and the number of losses ball in the match (r= 0.452). The largest coefficients of determination between the final place taken at the tournament and the implementation of technical and tactical actions were observed for those actions that were conventionally classified as actions of a mainly productive nature (for men - 0.69, for women - 0.73), and the smallest for actions mainly of a spectacular nature (for men and women - 0.33).

Torsion Angles between Donor and Acceptor Moieties as a Descriptor for Designing Nonlinear Optics and Thermally Activated Delayed Fluorescence Materials.

The performances of nonlinear optics (NLO) and thermally activated delayed fluorescence (TADF) materials are strongly related to the torsion angles (θ) between donor (D) and acceptor (A) moieties in D-A architecture molecules. However, the underlying relationships connecting θ to the performances of NLO/TADF materials remain unclear. Herein, we present a comprehensive theoretical study on NLO/TADF materials composed of a series of D-A backbone molecules (TPAAP/TPAAQ series and AQ-DMAC/AQ-MeFAC series) to shed light on these relationships. It is found that changing θ via the intramolecular locking strategy can greatly influence values of the first hyperpolarizability (β) and singlet-triplet energy gap (ΔEST), further leading to better/worse performances of NLO/TADF materials, respectively. Intriguingly, a more detailed analysis indicates that the variation trends between θ and β/ΔEST are changeable in low θ regions, exhibiting volcano-like relationships. The large coefficients of determination (R2, ranging from 0.76 to 0.93) suggest that this experimentally measurable parameter (θ) can be used as a promising descriptor to evaluate the performances of related materials. Following the revealed θ-β/θ-ΔEST correlations, the optimal/worst torsion angles for different materials are identified. These findings highlight the importance of the intrinsic structure-performance relationships, thus providing novel design strategies for high-performance NLO/TADF materials.

Assessing project portfolio risk via an enhanced GA-BPNN combined with PCA

Assessing project portfolio risk (PPR) is essential for organizations to grasp the overall risk levels of project portfolios (PPs) and realize PPR mitigation. However, current research is inadequate to effectively assess PPR, which brings challenges to managing PPR. In this context, the purpose of this study is to develop a PPR assessment model via an enhanced backpropagation neural network (BPNN). First, PPR assessment criteria considering project interdependencies are determined. Second, fuzzy logic is used to obtain original data for assessment criteria. Principal component analysis (PCA) is then employed to reduce the dimensionality of assessment criteria and derive the input and output of BPNN. Third, an improved genetic algorithm (IGA) is designed to optimize the initial weights and thresholds of BPNN. On this basis, the PCA-IGA-BPNN assessment model is constructed, followed by training and testing, possessing a test accuracy of 98.6%. Finally, comparison experiments are conducted from both internal and external perspectives. For internal comparison, the proposed model yields less mean absolute percentage error (MAPE), mean square error (MSE), and root mean square error (RMSE) than PCA-GA-BPNN, IGA-BPNN, PCA-BPNN and BPNN and offers the largest convergence speed (γ). As for external comparison, the presented model produces lower MAPE, MSE, and RMSE than Decision Tree (DT), Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbor (KNN) and has the largest coefficient of determination (R2). Results indicate that the established model performs more satisfactorily in assessing PPR. This research enriches PPR assessment methods and provides managers with a useful tool to evaluate PPR.

GROWTH CURVE OF A TROPICAL MIXED-GRASS PASTURE IN A HOT AND HUMID CLIMATE

<p><strong>Background. </strong>Pasture growth rate (PGR, kg/ha/day) depends on climatic and management practices. However, studies on the influence of the environment on pasture production and productivity of dry matter are scarce in tropical, hot, and humid regions of México. <strong>Objective.</strong> To estimate the pasture growth curve using time and climatic variables. <strong>Methodology.</strong> We related, through nonlinear models, the accumulated growth of a pasture composed of native grasses mixed with exotic grasses, using time and the variables temperature and day length as independent variables, the latter integrated into a single variable called thermal photo units (PTU). We estimated the daily growth rates of five divisions; from these, the forage yields for ten days until completing 29 periods. The best-fit models had the largest coefficients of determination and the lowest Akaike’s information criterion. <strong>Results.</strong> The model that best described the relationship between cumulative yield (Y) and cumulative growing days (X) was reciprocal-quadratic: y = x/(0.097535 – 0.0000881x + 0.0000006810x<sup>2</sup>) with R<sup>2</sup><sub>Adj,</sub> of 0.9988 and an AIC<sub>C</sub> of 222.6. The model that best described the relationship between the accumulated performance and the accumulated PTU was rational: y = (- 317.8 + 1.594x + 0.00001307x<sup>2</sup>)/(1 + 0.001059x + 0.00000001964x<sup>2</sup>), with R<sup>2</sup><sub>Adjt</sub>.=0.9985 and AIC<sub>C</sub>=233.4. Likewise, a two-segment model showed a close fit. The logarithmic model described the first segment: y<sub>1</sub> = -2268 + 417.2*(ln(x)), when y<sub>2</sub> =1079.3e<sup>0.00003932X</sup>, if x > 8415; with R<sup>2</sup><sub>Adj.</sub> = 0.9975 and AICc = 245.1. The value 8415 PTU was when the first derivative of both models coincided. <strong>Implications.</strong> The information generated is useful because it allows grazing system adjustment concerning the correct stocking rate application and designing more efficient grazing rotations. <strong>Conclusions. </strong>The conversion of growth rates to accumulated yield for ten-day periods produced a smooth curve that allowed fitting high-precision nonlinear models to predict forage accumulation from time and climatic variables.</p>

PENERAPAN METODE GEOGRAPHICALLY WEIGHTED REGRESSION PADA TINGKAT PENCEMARAN AIR BERDASARKAN TOTAL COLIFORM DI PROVINSI JAMBI

The purpose of this study is to obtain a local estimation model formed by considering spatial factors on the level of water pollution based on total coliforms in Jambi Province and can determine the most dominant factors affecting the total coliform water of PDAM Districts / Municipalities in Jambi Province using Geographically Weighted Regression (GWR) analysis. The type of data used in this study is primary data in quantitative form. This primary data was obtained by taking PDAM water samples directly in 9 regencies/cities in Jambi Province. The results of GWR analysis show that the dominant factors affecting the total coliform of PDAM water in each district/city in Jambi Province vary with different estimated parameters. The best model formed is a model with the GWR method, the adaptive bisquare weighting function which is shown from the value of the model's goodness criteria, namely the smallest AIC value and the largest coefficient of determination

FORECASTING MODEL OF ONIONS IN SUMBAWA DISTRICT

Sumbawa Regency as the second largest shallot producing area in NTB certainly contributes to food security in Sumbawa Regency in particular and in Indonesia in general. This condition certainly needs to make policy makers predict crop yield growth for the following years. This study aims to predict shallot yields for the next 9 years. The data used is secondary data sourced from the Sumbawa District Agriculture Office. There are three trend forecasting methods used, namely least square method, quadratic and exponential trend models. Based on the calculation results, the best forecasting trend model is obtained, namely the exponential trend model with MAPE and MAD values ​​and the largest coefficient of determination (R2). The exponential trend obtained shows a positive trend, namely positive exponential values ​​and positive principal numbers

QUANTILE REGRESSION MODEL ON RAINFALL IN MAKASSAR 2019

Makassar is an area that has a monsoon rainfall pattern. This study aims to find a quantile regression model and to determine the factors that significantly influence rainfall in the city of Makassar. This applied research applies a quantile regression model to rainfall data which is seasonal data. The advantage of this quantile regression model is that it is able to detect extreme conditions of rainfall, such as heavy rain. The data used is daily data in 2019. The estimation results obtained 9 (nine) models from each quantile used. The best model is obtained based on the largest coefficient of determination ( ), namely the 0,8th quantile ( ) of 0.28%. Furthermore, based on the model, it is found that the factor that significantly influences rainfall in the city of Makassar is humidity. At the same time, the air temperature and wind speed have no significant effect on rainfall in the city of Makassar.

Novel ensemble modelling for prediction of fundamental properties of bitumen incorporating plastic waste

Plastic asphalt mixtures (PAMs) have garnered attention recently, but their field application has been limited due to a lack of understanding of asphalt mix behavior following modification. A modelling tool that can calculate the plastic influence on the characteristics of asphalt mixtures is required to close this gap. Hence, this study offers a performance analysis of various machine learning (ML) models in predicting the performance of PAMs through its various properties. These models include three methods, decision tree (DT) as an individual technique, adaboost regressor (AR), and bagging regressor (BR), as ensemble techniques for prediction of fundamental properties of PAMs i.e. air voids (Va), marshall flow (MF), marshall stability (MS), tensile strength ratio (TSR), and indirect tensile strength (ITS). A series of experimental works and their results on the PAMs properties were collected through literature, to develop ML models and compare their accuracy. The comparative findings demonstrated that the BR model had the largest coefficient of determination (R2) and the lowest statistical errors, making it the model with the best predictive ability. According to the study's findings, ensemble approaches can be effectively utilized to forecast different properties of PAMs. Comparing ensemble models to the single model that served as their base learner, ensemble models were able to increase prediction accuracy. Furthermore, cross validation was performed to check the accuracy of developed models. SHAP analysis was conducted to examine the effects of input parameters on the fundamental properties of PAMs.

Development and validation of the nutrient-rich food index for China

To select the nutrient-rich food(NRF) index applicable to the food characteristics of Chinese residents through population-based dietary quality validation. Sixteen NRF indices were constructed based on the Chinese Food Composition Table and the Chinese Food Labelling Nutrient Reference Values, using 100 g as the base amount of food for calculation. The NRF indices were validated by the China Prime Diet Quality Score(CPDQS), a dietary quality assessment index, using food intake data collected from 14 068 adult residents in 15 provinces in 2018 under the China Health and Nutrition Survey. The CPDQS was used as the dependent variable and the dietary NRF index score was used as the independent variable. The multiple linear regression models were performed after adjusting the covariates and the NRF index with the largest coefficient of determination R~2 was used as the most appropriate index. Regression analysis showed that the R~2 of NRF8.2 index was the largest, which was 0.16. Dried legumes, fungi and algae, nuts, vegetables, eggs and aquatic products scored higher on the NRF8.2 index. The high scoring subgroups had higher average daily intakes of cereals and potatoes, vegetables, fruits, soybeans and nuts, dairy, eggs and aquatic products, as well as lower consumption of livestock and poultry meat, beverages, salt and cooking oil, compared to the low scoring subgroup on the Dietary NRF8.2 Index. In terms of nutrient intake, the medium and high NRF8.2 subgroups had relatively high intakes of energy, protein, vitamins and minerals and relatively low intakes of fat and sodium. The NRF8.2 index explains the maximum variation in CPDQS and is a relatively suitable NRF index for the Chinese population.

Two-parameters single equation stem taper models of Pinus sibirica in Siberia, Russia

A large number of studies have been devoted to the cedar forests in Russia, but studies of the conicity of trunks remain limited. The main goal of this study was to analyze two-parameter single equation stem tapper models for Siberian cedar trees, which will serve as guidelines for future research. The study tested six two-parameter models of one equation for the conicity of the trunk of Siberian cedar trees. For this, data from 1805 measurements of diameters in 193 felled trees were used. The data were collected in natural pure and mixed Siberian pine stands in Irkutsk region (Southern Siberia) and Tyumen region (Western Siberia), Russia. Our analysis showed that for our data set, the Reed and Green model turned out to be the best model. The model is characterized by the smallest average absolute error percentage, the square root of the root mean square error, and the largest coefficient of determination. The residuals of this model are independent and characterized by constant variance. The analysis performed showed that two-parameter single equation models have a good fit in the central part of the trunk, therefore, forestry management in cedar forests should be used with caution. For the purposes of forestry management in cedar forests, it is recommended to develop and put into practice more complex models.

SIMULTANEOUS SPATIAL DURBIN FOR PANEL DATA WITH TWO WAY ERROR COMPONENT MODEL

This study developed a Simultaneous Spatial model for panel data using Simultaneous Spatial Durbin Model for panel data with two way error component by using Generalized Spatial Two Stage Least Square (SSDPDTWEC GS2SLS). Based on the results of the simulation data, it can be concluded that the SSDPDTWEC GS2SLS model is better than Simultaneous Spatial Durbin for panel data with one way error component using Spatial Two Stage Least Square (SSDPDONWEC S2SLS), SSDPDONWEC GS2SLS, and SSDPDTWEC S2SLS, because it has a smaller mean of square error (MSE) and has a larger coefficient of determination (R Square) value.
 Based on the results of the SSDPDTWEC GS2SLS Model to endogenous economics growth data in Indonesia from 2015-2020, it can be concluded that Gross Regional Domestics Product (GRDP) is significantly affected by Human Development Index (HDI), Poverty (POOR), spatial dependencies of the HDI (WHDI), spatial dependencies of poverty (WPOOR) and spatial dependencies of GRDP (WGRDP), while Information and Communication Technology Development Index (ICTDI) and spatial dependencies of ICTDI (WICTDI) have no significant effect on GRDP. For ICTDI is significantly affected by HDI, WHDI and WICTDI, while GRDP, Direct Expenditure (DE), WGRDP and spatial dependencies of direct expenditure (WDE) have no significant effect on ICTDI. HDI variable is significantly affected by ICTDI, WICTDI, and WHDI while GRDP and WGRDP have no significant effect on HDI. There is no spatial simultaneity relationship between GRDP and HDI as well as GRDP and ICTDI, while between ICTDI and HDI there is a spatial simultaneous relationship.

COMPETITIVE ADVANTAGE TOURISM BASED ON E-TOURISM TO IMPROVE TOURIST DESTINATIONS

This study aims to determine the influence of the determining component of Competitive Advantage Tourism in increasing tourist destinations based E Tourism. This research was conducted in Kerinci Regency, Jambi Province. The population in this study was (1) Village Government or Local Government (2) Hotel and lodging Managers (3) Restaurant and Restaurant Managers (4) Travel Agency Managers (5) Managers at Attractions (6) Tourism Services Support (7) Local People in tourism destination obtained a total sample of 120 respondents. The data used in this study were primary data through interviews and questionnaires. Data analysis in this study uses a structural equation model with the analysis tool used is to use AMOS. The results of the study obtained a large coefficient of determination of the influence of Competitive Advantage Tourism which can be explained by the variables Built Environment, Operating Sectors, Natural Resources Environment, and Spirit of Hospitality Cultural and E Tourism of 81.4%. While the coefficient of determination of the Tourism Destination equation is 1.004 or the Tourism Destination variable which can be explained by the variability of Tourism Destination and E Tourism is 100%

Quantitative analysis of polycyclic aromatic hydrocarbons (PAHs) in water by surface-enhanced Raman spectroscopy (SERS) combined with Random Forest

Polycyclic aromatic hydrocarbons (PAHs) have strong carcinogenicity, teratogenicity, mutagenicity and other adverse effects on human beings. They are one of the most dangerous pollutants, which have attracted great attention in the past decades. In this work, aiming at the actual problems that water environment is polluted and human health is threatened by PAHs, surface enhanced Raman spectroscopy (SERS) combined with Random Forest (RF) calibration models were used to quantitative analysis of phenanthrene and fluoranthene in water. Firstly, the SERS data was collected after samples mixed with Ag NPs, after 31 PAHs samples were prepared. Secondly, it was discussed how spectral preprocessing integration strategies affect on the prediction performance of the RF calibration models. And then, the effect of mutual information (MI) variable selection method on the performance of RF calibration models was explored. Finally, the RF calibration models were established for phenanthrene and fluoranthene. For the prediction set, a lowest mean relative error (MRE) and a largest determination coefficient (R2) were obtained. For quantitative analysis of phenanthrene, the final prediction performance results show that R2p is 0.9780, and MREp is 0.0369 based on the D1st-WT-RF calibration model. For fluoranthene, WT-D1st-MI-RF is a better calibration model, and corresponding to R2p and MREp are 0.9770 and 0.0694, respectively. Hence, a rapid and accurate quantitative method of PAHs is established for the real-time detection of water environmental pollution, which is intended to provide new ideas and methods for the quantitative analysis of PAHs in water.

Numerical simulation for porous hemp plastic composites sound absorption properties

Hemp is an environmentally friendly porous fiber, and has good sound absorption properties, which can be used to reduce noise. In recent years, hemp-based composites have been widely studied, but most of them study the preparation methods of materials, Therefore, this paper studies the topic. Firstly, the material and specimens with different thickness and diameter were prepared, then, the standing wave tube method experiment was carried out, and the values of sound absorption coefficient of the specimens were measured and analyzed, while a novel model using genetic algorithm and artificial neural network is proposed. Secondly, the artificial neural network structure was designed, and the training data, verification data, test data are divided and the data is preprocessed. Thirdly, improved genetic algorithm is proposed to design to determine an optimal solution. Fourthly, the optimal solution was as the initial weight and threshold, and were input into the artificial neural network together with training data and verification data. Levenberg Marquardt algorithm were used to train the network. Finally, input the test data into the trained network to test the model. The results showed that there is only a minor deviation between the model output values and the measured values, the values of sound absorption coefficient of specimens with 10 mm thickness are between 0.9 and 1. It was also found that compared with artificial neural network model without genetic algorithm, the proposed model has the lower root mean squared error (RMSE) value, the larger coefficient of determination (R2) value.

Electromagnetic characteristic estimation on spiral antennas through AOI, ML, and AI

In this study, a method that is able to estimate the electromagnetic characteristic of spiral antennas was proposed and realized through consecutive procedures of automatic optical inspection (AOI), machine learning (ML), and artificial intelligence (AI), providing a solution to smart manufacturing. Two-arm self-complementary Archimedean spiral antennas (SCASAs) were introduced as examination targets with pattern distortions from potential process variations, in which bulges and neckings were mathematically generated to imitate uncontrollable ink rheology in printed and flexible electronics, covering the unexplored parts in previous works. The SCASAs in the training group were fabricated by standard printed circuit board procedures, and their pattern integrity in terms of line edge roughness (LER) and coupling frequency were collected through AOI for ML as the feature and label, respectively. The established AI model was based on Gaussian process regression with covariance function of exponential that showed the smallest root-mean-square-error and the largest coefficient of determination through iterative lazy-learning. By feeding the LERs of the SCASAs into the testing group, their corresponding coupling frequencies were estimated by the established AI model with high confidence level. Good linearity between the estimated and measured responses indicated that a reliable AI model and procedure were built, which outperforms existing methods that are unable to project off-line active characteristics of microelectronic components from their in-line pattern integrities.

Machine learning and Python assisted design and verification of Fe–based amorphous/nanocrystalline alloy

We report a machine learning (ML) and Python assisted strategy to accelerate the design and verification of Fe–based amorphous and nanocrystalline alloy with desired properties. Linear Regression (LR), Support Vector Regression (SVR), Decision Tree Regression (DTR), Artificial Neural Network (ANN) and Random Forest Regression (RFR) are employed to build prediction models of soft magnetic properties, such as saturation magnetic flux density (Bs), coercivity force (Hc), magnetization (Ms), Curie temperature (Tc), maximum permeability (µmax) and effective permeability (µe). It is found that ANN has the excellent fitting ability with largest coefficient of determination (R2) to predict the soft magnetic properties of new designed alloys. Then, Python screening is used to find the alloy compositions with best soft magnetic properties of Fe–B–P–C–Nb system. Finally, Fe83B9P3C4Nb1 alloy with good soft magnetic properties has been designed and prepared to verify. It is indicated that the soft magnetic properties of Fe83B9P3C4Nb1 amorphous and nanocrystalline alloy predicted by ML are in agreement with the experimental measured results. These findings indicate that ML and Python assisted approach can accelerate the design of Fe–based alloys with desired properties accurately.