Least Angle Regression Research Articles

Study on detection of pesticide residues in tobacco based on hyperspectral imaging technology.

Tobacco is a critical economic crop, yet its cultivation heavily relies on chemical pesticides, posing health risks to consumers, therefore, monitoring pesticide residues in tobacco is conducive to ensuring food safety. However, most current research on pesticide residue detection in tobacco relies on traditional chemical methods, which cannot meet the requirements for real-time and rapid detection. This study introduces an advanced method that combines hyperspectral imaging (HSI) technology with machine learning algorithms. Firstly, a hyperspectral imager was used to obtain spectral data of tobacco samples, and a variety of spectral pre-processing technologies such as mean centralization (MC), trend correction (TC), and wavelet transform (WT), as well as feature extraction methods such as competitive adaptive reweighted sampling (CARS) and least angle regression (LAR) were used to process the spectral data, and then, grid search algorithm (GSA) is used to optimize the support sector machine (SVM). The optimized MC-LAR-SVM model achieved a pesticide classification accuracy of 84.1%, which was 9.5% higher than the original data model. The accuracy of the WT-TC-CARS-GSA-SVM model in the fenvalerate concentration classification experiment was as high as 91.8 %, and it also had excellent performance in other metrics. Compared with the model based on the original data, the accuracy, precision, recall, and F1-score are improved by 8.3 %, 8.2 %, 7.5 %, and 0.08, respectively. The results show that combining spectral preprocessing and feature extraction algorithms with machine learning models can significantly enhance the performance of pesticide residue detection models and provide robust, efficient, and accurate solutions for food safety monitoring. This study provides a new technical means for the detection of pesticide residues in tobacco, which is of great significance for improving the efficiency and accuracy of food safety detection.

BiLSTM-InceptionV3-Transformer-fully-connected model for short-term wind power forecasting

Accurate short-term wind power prediction can help grid managers optimize the transmission and storage of electricity from wind farms, avoiding energy wastage and unnecessary operating costs, and ensuring the stability and reliability of power supply. However, stochastic nature and uncertainty data pose challenges for accurate wind power prediction models. This study combines complementary ensemble empirical mode decomposition (CEEMD), discrete wavelet transform (DWT), bidirectional long short-term memory (BiLSTM), InceptionV3, Transformer, and fully connected neural networks for wind power prediction. The proposed prediction model can fully capture the nonlinear and dynamic characteristics of the wind speed data that are decomposed into intrinsic modal functions by CEEMD. Simultaneously, the prediction model can capture the local characteristics and cyclic variations of the data at different time scales based on fifth-order DWT. A dataset spanning 378 days from Natal is employed and the proposed model is compared with 19 other networks (e.g., ResNet, GoogleNet, SqueezeNet) and 15 other traditional regression algorithms (e.g., Gaussian process regression, least-angle regression, and nearest neighbors regression). The experimental results show that the designed BiLSTM-InceptionV3-Transformer-fully-connected model (BITFM) can reduce the mean absolute error by at least 38.75 % and the root mean squared error by at least 33.33 % compared with 19 other networks. Compared with 15 traditional regression prediction models, BITFM can reduce the mean absolute error by at least 57.41 % and the root mean squared error by at least 53.95 %. To further validate the prediction performance of BITFM, this study conducts tests employing data from the Santa Vitoria do Palmar. The experimental results indicate that BITFM consistently achieved the best performance metrics.

Identification and quantification of adulteration in collagen powder by terahertz spectroscopy − the effect of spectral characteristics on performance is considered

Terahertz spectroscopy is an emerging rapid detection method that can be used to detect and analyze food quality issues. However, models developed based on various spectral characteristics of terahertz have shown different performances in food identification. Therefore, we preliminarily analyzed the effect of terahertz spectral characteristics on the identification and quantification of collagen powder adulterated with food powders (plant protein powder, corn starch, wheat flour) with the use of random forest (RF), linear discriminant analysis (LDA), and partial least squares regression (PLSR), and determined the spectral characteristics suitable for identification and quantitative analysis. Then, the selected spectral characteristics data were preprocessed using baseline correction (BC), gaussian filter (GF), moving average (MA), and savitzky-golay (SG). Feature variables were extracted from preprocessed spectral characteristics data using genetic algorithm (GA), random forest (RF), and least angle regression (LAR). The study indicated that the BC-GA-LDA classification model based on the absorption coefficient spectra achieved an accuracy of 96.96% in identifying adulterated collagen powder. Additionally, the GA-PLSR model developed based on the power spectra demonstrated excellent performance in predicting adulteration levels, with the coefficient of determination (Rp2) values ranging from 0.93 to 0.99. The results showed that the rational selection of terahertz spectral characteristics is highly feasible for the accurate detection of collagen powder adulteration.

A machine learning approach for predicting flexural strength of 3D printed hexagon lattice-cored sandwich structures

The development of sandwich structures using Fused Filament Fabrication (FFF) technology is an effective method for the rapid construction of complex profiles and is widely recognized for various structural applications. In this study, hexagon lattice-cored sandwich structures are developed by placing the lattice core in the center portion of the PLA polymeric specimen. The performance is analyzed with respect to varying levels of 3D-Printing Factors (3D-PFs) such as Nozzle Temperature (NT), Line Width (LW), Printing Speed (PS), and Layer Height (LH). The levels of the respective 3D-PFs are varied as follows: NT (180, 190, 200, and 210 °C), LH (0.15, 0.2, 0.25, and 0.3 mm), PS (15, 20, 25, and 30 mm/s), and LW (0.1, 0.2, 0.3, and 0.4 mm). The sandwich flexural samples are 3D printed using an FFF 3D printer, and their flexural properties are examined using a Universal Testing Machine (UTM). In the present work, various Machine Learning (ML) algorithms, such as Least Angle Regression (LARS), Support Vector Machine (SVM), Decision Tree (DT), and Ridge Regression (Ridge), are employed to predict the Flexural Strength (FS) of the 3D-printed sandwich structures and help in determining the optimized levels of 3D-PFs for achieving higher FS. The findings indicate that the LARS algorithm, particularly when applied within the Bagging Ensemble Learning Technique (ELT), exhibits the highest precision, showcasing a Mean Absolute Error (MAE) of 0.163, Root Mean Square Error (RMSE) of 0.153, and Median Absolute Error (MedAE) of 0.129. With the help of the LARS algorithm under Bagging ELT, the impact of 3D-PFs on the resulting FS is analyzed, which helps in determining the optimized levels of 3D-PFs concerning FS. The sandwich structures fabricated at the optimized levels show improved flexural properties, making them suitable for various structural applications.

Multi-objective robust optimization design framework for low-pollution emission burners

The optimal design of low-pollution emission burners plays an important role in controlling pollutant emissions of industrial equipment, and is crucial for the sustainable development of the national economy and environmental protection. However, many uncertain factors challenge the optimal design of low-pollution emission burners. The Latin hypercube sampling (LHS) method was used to obtain sampling data representing the distribution of the uncertain variable. The training dataset was obtained using the turbulent combustion coupling model. A high-precision sparse polynomial chaos expansion (PCE) model was constructed by the degree-adaptive scheme and least angle regression (LAR) algorithm. Furthermore, the Legendre polynomial is introduced to establish a continuous robust optimization model. The model is carried out by the non-dominated sorting genetic algorithm II (NSGA-II). The results show that the excess air coefficient of 1.227 is optimal. Compared with the excess air coefficient of 1.20 under the discrete robust optimization, the optimal coefficient can further reduce pollutant emissions and bring strong robustness to the ethylene cracking furnace. It has also been proven that the continuous robust optimization scheme improves the optimization granularity. Compared with discrete robust optimization, this method reduces the number of samples by 66.7 %.

Differences of soil carbon pools and crop growth across different typical agricultural fields in China: The role of geochemistry and climate change

Carbon storage and the aboveground biomass of farmland provide practical significance for understanding global changes and ensuring food production and quality. Based on soil carbon storage, aboveground biomass, climate, geochemistry, and other data from 19 farmland ecological stations in China, we analysed the distribution characteristics of farmland carbon storage in topsoil and aboveground biomass. We notably revealed the response direction and degree of climate and geochemical factors to farmland carbon storage in topsoil and aboveground biomass. The results indicated that the average carbon stocks of farmland in different regions ranged from 0.28 to 7.91 kg m−2, the average fresh weight of the aboveground biomass (FAB) ranged from 1370.64 to 5997.28 g m−2, and the average dry weight of the aboveground biomass (DAB) ranged from 119.95 to 852.35 g m−2. The least angle regression (LARS) and the best subsection selection regression (BSS) showed that evapotranspiration and extreme low temperatures were significant climatic factors affecting carbon sequestration and aboveground biomass on long-time scales. The linear mixed-effects model (LMM) further showed that AN and AP had significant long-term effects on carbon sequestration and aboveground biomass (p < 0.05), with AN having the highest contribution to SOC%, FAB, and DAB. The structural equation model (SEM) showed that carbon sequestration and aboveground biomass in agricultural fields were significantly positively correlated (p < 0.05). Moreover, the climate had a less direct contribution to carbon sequestration and above-ground biomass compared to geochemistry (PCc < 0.1<PCG), and its effect was more indirect. When the geochemical variables were removed, the correlation between climate and carbon and aboveground biomass variables decreased significantly. We conclude that it is possible that extremes climate and geochemistry control carbon sequestration and above-ground biomass in long-term cropland through interactions in the context of global change, which provides new insights into the evolution of soil organic carbon in long-term cropland and the scientific formulation of policies to increase food production and preserve quality.

Least angle regression, relaxed lasso, and elastic net for algebraic multigrid of systems of elliptic partial differential equations

AbstractIn a sequence of papers, the author examined several statistical affinity measures for selecting the coarse degrees of freedom (CDOFs) or coarse nodes (Cnodes) in algebraic multigrid (AMG) for systems of elliptic partial differential equations (PDEs). These measures were applied to a set of relaxed vectors that exposes the problematic error components. Once the CDOFs are determined using any one of these measures, the interpolation operator is constructed in a bootstrap AMG (BAMG) procedure. However, in a recent paper of Kahl and Rottmann, the statistical least angle regression (LARS) method was utilized in the coarsening procedure and shown to be promising in the CDOF selection. This method is generally used in the statistics community to select the most relevant variables in constructing a parsimonious model for a very complicated and high‐dimensional model or data set (i.e., variable selection for a “reduced” model). As pointed out by Kahl and Rottmann, the LARS procedure has the ability to detect group relations between variables, which can be more useful than binary relations that are derived from strength‐of‐connection, or affinity measures, between pairs of variables. Moreover, by using an updated Cholesky factorization approach in the regression computation, the LARS procedure can be performed efficiently even when the original set of variables is large; and due to the LARS formulation itself (i.e., its ‐norm constraint), sparse interpolation operators can be generated. In this article, we extend the LARS coarsening approach to systems of PDEs. Furthermore, we incorporate some modifications to the LARS approach based on the so‐called elastic net and relaxed lasso methods, which are well known and thoroughly analyzed in the statistics community for ameliorating several major issues with LARS as a variable selection procedure. We note that the original LARS coarsening approach may have addressed some of these issues in similar or other ways but due to the limited details provided there, it is difficult to determine the extent of their similarities. Incorporating these modifications (or effecting them in similar ways) leads to improved robustness in the LARS coarsening procedure, and numerical experiments indicate that the changes lead to faster convergence in the multigrid method. Moreover, the relaxed lasso modification permits an indirect BAMG (iBAMG) extension to the interpolation operator. This iBAMG extension applied in an intra‐ or inter‐variable interpolation setting (i.e., nodal‐based coarsening), as well as in variable‐based coarsening, which will not preserve the nodal structure of a finest‐level discretization on the lower levels of the multilevel hierarchy, will be examined. For the variable‐based coarsening, because of the parsimonious feature of LARS, the performance is reasonably good when applied to systems of PDEs albeit at a substantial additional cost over a nodal‐based procedure.

Efficient MAPoD via Least Angle Regression based Polynomial Chaos Expansion Metamodel for Eddy Current NDT

In this article, a metamodeling approach based on non-intrusive polynomial chaos expansion (PCE) with least angle regression (LAR) method is used in boundary element analysis for a model-assisted probability of detection (MAPoD) study of eddy current nondestructive testing (NDT) systems. The LAR-PCE metamodel represents the NDT system model responses by a set of coefficients with the polynomial basis functions in lieu of pure kernel degeneration accelerated boundary element method (KD-BEM) based physical model. Both the computational accuracy and efficiency of the LAR-PCE metamodel over the ordinary least squares (OLS) based PCE metamodel are demonstrated by testing the 3D eddy current NDT benchmarks with different system setups, flaw lengths and widths. The simulation results show two digits accuracy of the PoD metrics compared with the ones achieved by the KD-BEM based physical model as the benchmark. The LAR-PCE metamodel has remarkable improvements in computational efficiency over the OLS-PCE metamodel and accelerates the MAPoD study.

Improved prediction of vitamin C and reducing sugar content in sweetpotatoes using hyperspectral imaging and LARS-enhanced LASSO variable selection

This study utilized least angle regression (LARS) with the least absolute shrinkage and selection operator (LASSO) to select important wavelengths for rapid quantification of vitamin C (Vc) and reducing sugar (RS) in sweetpotato roots (SPR) using hyperspectral imaging (900–1700 nm). Nine wavelengths strongly correlated with Vc levels and twelve with RS were identified, achieving good predictions with partial least squares (PLS) regression (Vc: rP = 0.9704, RMSEP = 1.0098 mg/100 g; RS: rP = 0.9641, RMSEP = 0.2725 g/100 g). Validation with an independent sample set (n = 35) showed minimal deviation between predicted and actual values (Vc: 1.179–1.211 mg/100 g; RS: 0.316–0.324 g/100 g). Explainable AI and SHapley Additive exPlanations (SHAP) values were used to interpret the selected wavelengths. Chemical maps visually analyzed Vc and RS distribution in different SPR samples. This method effectively estimates Vc and RS levels in SPR, potentially aiding SPR quality assessment during post-harvest marketing and storage.

Modeling of Effectiveness of N3-Substituted Amidrazone Derivatives as Potential Agents against Gram-Positive Bacteria.

Prediction of the antibacterial activity of new chemical compounds is an important task, due to the growing problem of bacterial drug resistance. Generalized linear models (GLMs) were created using 85 amidrazone derivatives based on the results of antimicrobial activity tests, determined as the minimum inhibitory concentration (MIC) against Gram-positive bacteria: Staphylococcus aureus, Enterococcus faecalis, Micrococcus luteus, Nocardia corallina, and Mycobacterium smegmatis. For the analysis of compounds characterized by experimentally measured MIC values, we included physicochemical properties (e.g., molecular weight, number of hydrogen donors and acceptors, topological polar surface area, compound percentages of carbon, nitrogen, and oxygen, melting points, and lipophilicity) as potential predictors. The presence of R1 and R2 substituents, as well as interactions between melting temperature and R1 or R2 substituents, were also considered. The set of potential predictors also included possible biological effects (e.g., antibacterial, antituberculotic) of tested compounds calculated with the PASS (Prediction of Activity Spectra for Substances) program. Using GLMs with least absolute shrinkage and selection (LASSO), least-angle regression, and stepwise selection, statistically significant models with the optimal value of the adjusted determination coefficient and of seven fit criteria were chosen, e.g., Akaike's information criterion. The most often selected variables were as follows: molecular weight, PASS_antieczematic, PASS_anti-inflam, squared melting temperature, PASS_antitumor, and experimental lipophilicity. Additionally, relevant to the bacterial strain, the interactions between melting temperature and R1 or R2 substituents were selected, indicating that the relationship between MIC and melting temperature depends on the type of R1 or R2 substituent.

Multi-source uncertainty propagation and sensitivity analysis of turbine blades with underplatform dampers

Underplatform dampers (UPDs) mitigate turbine blade vibrations in aeroengines through friction dissipation generated by the contact interface. However, in UPD design, uncertainties are often overlooked, including manufacturing discrepancies, excitation forces, and wear factors, leading to suboptimal predictions of structural dynamic responses. This study presents a dynamic model for the blade-UPD system with cyclic symmetric attributes, which simulates uncertainties using statistical methods. An efficient algorithm using adaptive techniques is proposed to construct polynomial chaos expansions (PCE) for precise and efficient uncertainty quantification (UQ) in turbine blades with UPDs. Further, the influence of single/multiple parameter uncertainties on the dynamic characteristics of the blade-UPD is explored. Sobol' indices are then employed to assess the sensitivity of uncertain factors to the vibration reduction properties of UPDs. The findings suggest that the new approach, which offers precise UQ at minimal computational cost, outperforms traditional methods like Ordinary Least Squares (OLS) and Sparse Least Angle Regression (LARS). Observations reveal a significant impact of parameter uncertainties on blade-UPD dynamic responses, which manifest as "resonance bands" and "frequency shifts" in some cases. Sensitivity analysis indicates noticeable variations in Sobol' indices for each uncertainty parameter as the excitation frequency changes. Specifically, the uncertainty in the friction coefficient demonstrates pronounced sensitivity to amplitude when slip occurs at the contact interface. Furthermore, the observed "drop" phenomenon in Sobol' indices is explained.

G–LIME: Statistical Learning for Local Interpretations of Deep Neural Networks Using Global Priors (Abstract Reprint)

To explain the prediction result of a Deep Neural Network (DNN) model based on a given sample, LIME [1] and its derivatives have been proposed to approximate the local behavior of the DNN model around the data point via linear surrogates. Though these algorithms interpret the DNN by finding the key features used for classification, the random interpolations used by LIME would perturb the explanation result and cause the instability and inconsistency between repetitions of LIME computations. To tackle this issue, we propose G-LIME that extends the vanilla LIME through high-dimensional Bayesian linear regression using the sparsity and informative global priors. Specifically, with a dataset representing the population of samples (e.g., the training set), G-LIME first pursues the global explanation of the DNN model using the whole dataset. Then, with a new data point, -LIME incorporates an modified estimator of ElasticNet-alike to refine the local explanation result through balancing the distance to the global explanation and the sparsity/feature selection in the explanation. Finally, G-LIME uses Least Angle Regression (LARS) and retrieves the solution path of a modified ElasticNet under varying -regularization, to screen and rank the importance of features [2] as the explanation result. Through extensive experiments on real world tasks, we show that the proposed method yields more stable, consistent, and accurate results compared to LIME.

Model order reduction based on low-rank approximation for parameterized eigenvalue problems in structural dynamics

This study introduces a novel reduced-order modeling approach for parameterized eigenvalue problems. The proposed framework represents eigenvectors as a low-rank approximation and is implemented using an offline-online strategy. In the offline stage, nonlinear equations are formulated by imposing a stationary condition on the weighted Rayleigh quotient over sampled points and eigenvalues of interest. A variant of the Krylov subspace approach is applied to progressively identify the dominant response characteristics at sampled points. In the online stage, a reduced-order model is obtained through Galerkin projection, and eigensolutions at unsampled points are determined. The performance of the proposed framework is validated through numerical examples, where perturbation, least-angle regression, and Gaussian process are considered for comparison. The obtained results demonstrate that the proposed framework can accurately estimate eigensolutions, even when eigenvalues are adjacent to each other. In terms of efficiency, the a priori construction of the low-dimensional basis is possible without solving the full-order model, resulting in substantial CPU time savings during the offline computation process. Lastly, the applicability of the proposed framework is investigated through an uncertainty quantification example using numerical and experimental results.

Pendugaan Koefisien Regresi Logistik Biner Menggunakan Algoritma Least Angle Regression

Binary logistic regression is a statistical analysis method that aims to determine the relationship between variable which has two categories with the predictor variable that have categorical or continuous scale. The method that used to estimate logistic regression parameters is Maximum Likelihood Estimation (MLE) method. In estimating parameters, Least Angle Regression (LAR) algorithm is used to select the significant variables in order to get the best model from the estimation results of binary logistic regression coefficients. This LAR algorithm is applied to the risko of stunting data in two-year-old-babies at Buntu Batu Health Center working area, Enrekang Regency, South Sulawesi in 2019. This results obtained in the estimation of binary logistic regression prediction model using LAR algorithm, the standard error value is 0.018 smaller than the standard error value of binary logistic regression, which is 0.025. This shows that the binary logistic regression model using LAR algorithm is better than the usual binary logistic regression model on the risk of stunting data. Based on the results obtained, the variables that significantly affect the risk of stunting in two-year-old-babies on 2019 are father’s height, body length of birth, exclusive breastfeeding, history of infectious diseases, and history of immunization.

Determination of soluble solids content of multiple varieties of tomatoes by full transmission visible-near infrared spectroscopy.

Soluble solids content (SSC) is a pivotal parameter for assessing tomato quality. Traditional measurement methods are both destructive and time-consuming. To enhance accuracy and efficiency in SSC assessment, this study employs full transmission visible and near-infrared (Vis-NIR) spectroscopy and multi-point spectral data collection techniques to quantitatively analyze SSC in two tomato varieties ('Provence' and 'Jingcai No.8' tomatoes). Preprocessing of the multi-point spectra is carried out using a weighted averaging approach, aimed at noise reduction, signal-to-noise ratio improvement, and overall data quality enhancement. Taking into account the potential influence of various detection orientations and preprocessing methods on model outcomes, we investigate the combination of partial least squares regression (PLSR) with two orientations (O1 and O2) and two preprocessing techniques (Savitzky-Golay smoothing (SG) and Standard Normal Variate transformation (SNV)) in the development of SSC prediction models. The model achieved the best results in the O2 orientation and SNV pretreatment as follows: 'Provence' tomato (Rp = 0.81, RMSEP = 0.69°Brix) and 'Jingcai No.8' tomatoes (Rp = 0.84, RMSEP = 0.64°Brix). To further optimize the model, characteristic wavelength selection is introduced through Least Angle Regression (LARS) with L1 and L2 regularization. Notably, when λ=0.004, LARS-L1 produces superior results ('Provence' tomato: Rp = 0.95, RMSEP = 0.35°Brix; 'Jingcai No.8' tomato: Rp = 0.96, RMSEP = 0.33°Brix). This study underscores the effectiveness of full transmission Vis-NIR spectroscopy in predicting SSC in different tomato varieties, offering a viable method for accurate and swift SSC assessment in tomatoes.

Characteristic Analysis and Short-Impending Prediction of Aircraft Bumpiness over Airport Approach Areas and Flight Routes

Based on the Quick Access Recorder (QAR) data covering over 9000 routes in China, the monthly and intra-day distribution characteristics of aircraft bumpiness at different levels were analyzed, and the relationships between the eddy dissipation rate (EDR) and other aircraft flight status elements during bumpiness occurrence were also analyzed. Afterward, aircraft bumpiness routes were constructed using 19 machine learning models. The analyses show that (1) aircraft bumpiness was mainly concentrated between 0:00 a.m. and 17:00 p.m. Severe aircraft bumpiness occurred more frequently in the early morning in January, especially between 5:00 a.m. and 6:00 a.m., and moderate bumpiness always occurred from 3:00 a.m. to 11:00 a.m. (2) The relationship between the left and right attack angles and aircraft bumpiness on the routes was more symmetrical, with a center at 0 degrees, unlike in the approach area where the hotspots were mainly concentrated in the range of −5 to 0 degrees. In the approach area, the larger the Mach number, the more severe the bumpiness. (3) The performances of the Automatic Relevance Determination Regression (ARD), Partial Least Squares Regression (PLS), Elastic-Net Regression (ENR), Classification and Regression Tree (CART), Passive Aggressive Regression (PAR), Random Forest (RF), Stochastic Gradient Descent Regression (SGD), and Tweedie Regression (TWD) based models were relatively good, while the performances of the Huber Regression (HUB), Least Angle Regression (LAR), Polynomial Regression (PLN), and Ridge Regressor (RR) based models were very poor. The aircraft bumpiness prediction models performed best over the approach area of ZBDT (airport in Datong), ZULS (airport in Lhasa), ZPPP (airport in Kunming), and ZLQY (airport in Qingyang). The model performed best in predicting the ZLLL-ZBDT air route (flight routes for Lanzhou to Datong) with different prediction times.

NIR Spectral Inversion of Soil Physicochemical Properties in Tea Plantations under Different Particle Size States.

Soil fertility is vital for the growth of tea plants. The physicochemical properties of soil play a key role in the evaluation of soil fertility. Thus, realizing the rapid and accurate detection of soil physicochemical properties is of great significance for promoting the development of precision agriculture in tea plantations. In recent years, spectral data have become an important tool for the non-destructive testing of soil physicochemical properties. In this study, a support vector regression (SVR) model was constructed to model the hydrolyzed nitrogen, available potassium, and effective phosphorus in tea plantation soils of different grain sizes. Then, the successful projections algorithm (SPA) and least-angle regression (LAR) and bootstrapping soft shrinkage (BOSS) variable importance screening methods were used to optimize the variables in the soil physicochemical properties. The findings demonstrated that soil particle sizes of 0.25-0.5 mm produced the best predictions for all three physicochemical properties. After further using the dimensionality reduction approach, the LAR algorithm (R2C = 0.979, R2P = 0.976, RPD = 6.613) performed optimally in the prediction model for hydrolytic nitrogen at a soil particle size of 0.25~0.5. The models using data dimensionality reduction and those that used the BOSS method to estimate available potassium (R2C = 0.977, R2P = 0.981, RPD = 7.222) and effective phosphorus (R2C = 0.969, R2P = 0.964, RPD = 5.163) had the best accuracy. In order to offer a reference for the accurate detection of soil physicochemical properties in tea plantations, this study investigated the modeling effect of each physicochemical property under various soil particle sizes and integrated the regression model with various downscaling strategies.

Multiple Linear Regression and Machine Learning for Predicting the Drinking Water Quality Index in Al-Seine Lake

Ensuring safe and clean drinking water for communities is crucial, and necessitates effective tools to monitor and predict water quality due to challenges from population growth, industrial activities, and environmental pollution. This paper evaluates the performance of multiple linear regression (MLR) and nineteen machine learning (ML) models, including algorithms based on regression, decision tree, and boosting. Models include linear regression (LR), least angle regression (LAR), Bayesian ridge chain (BR), ridge regression (Ridge), k-nearest neighbor regression (K-NN), extra tree regression (ET), and extreme gradient boosting (XGBoost). The research’s objective is to estimate the surface water quality of Al-Seine Lake in Lattakia governorate using the MLR and ML models. We used water quality data from the drinking water lake of Lattakia City, Syria, during years 2021–2022 to determine the water quality index (WQI). The predictive performance of both the MLR and ML models was evaluated using statistical methods such as the coefficient of determination (R2) and the root mean square error (RMSE) to estimate their efficiency. The results indicated that the MLR model and three of the ML models, namely linear regression (LR), least angle regression (LAR), and Bayesian ridge chain (BR), performed well in predicting the WQI. The MLR model had an R2 of 0.999 and an RMSE of 0.149, while the three ML models had an R2 of 1.0 and an RMSE of approximately 0.0. These results support using both MLR and ML models for predicting the WQI with very high accuracy, which will contribute to improving water quality management.

Research on Apple Origins Classification Optimization Based on Least-Angle Regression in Instance Selection

Machine learning is used widely in near-infrared spectroscopy (NIRS) for fruit qualification. However, the directly split training set used contains redundant samples, and errors may be introduced into the model. Euclidean distance-based and K-nearest neighbor-based instance selection (IS) methods are widely used to remove useless samples because of their accessibility. However, they either have high accuracy and low compression or vice versa. To compress the sample size while improving the accuracy, the least-angle regression (LAR) method was proposed for classification instance selection, and a discrimination experiment was conducted on a total of four origins of 952 apples. The sample sets were split into the raw training set and testing set; the optimal training samples were selected using the LAR-based instance selection (LARIS) method, and the four other selection methods were compared. The results showed that 26.9% of the raw training samples were selected using LARIS, and the model based on these training samples had the highest accuracy. Thus, the apple origin classification model based on LARIS can achieve the goal of high accuracy and compression and provide experimental support for the least-angle regression algorithm in classification instance selection.

Development of explainable AI-based predictive models for bubbling fluidised bed gasification process

In this study, seven different types of regression-based predictive modelling techniques are used to predict the product gas composition (H2, CO, CO2, CH4) and gas yield (GY) during the gasification of biomass in a fluidised bed reactor. The performance of different regression-based models is compared with the gradient boosting model (GB) to show the relative merits and demerits of the technique. Additionally, SHapley Additive exPlanations (SHAP)-based explainable artificial intelligence (XAI) method was utilised to explain individual predictions. This study demonstrates that the prediction performance of the GB algorithm was the best among other regression-based models i.e. Linear Regression (LR), Multilayer perception (MLP), Ridge Regression (RR), Least-angle regression (LARS), Random Forest (RF) and Bagging (BAG). It was found that at learning rate (lr) 0.01 and number of boosting stages (est) 1000 yielded the best result with an average root mean squared error (RMSE) of 0.0597 for all outputs. The outcome of this study indicates that XAI-based methodology can be used as a viable alternative modelling paradigm in predicting the performance of a fluidised bed gasifier for an informed decision-making process.