Sequential Backward Selection Research Articles

A Machine Learning-Based Wrapper Method for Feature Selection

This paper presents a two-stage feature selection scheme using machine learning techniques. In the first stage a wrapper method is adopted to select various combinations of subsets of features from the original dataset. The performance of the model is evaluated by three classifiers: K-Nearest Neighbor (KNN), Support Vector Machines (SVM), and Random Forest (RF). In the second and final stage, a sequential backward feature selection Method is applied. The proposed method is demonstrated on eighteen datasets and the average classification accuracy of eighteen datasets achieved is 89.81%, 87.55%, and 89.82% using the KNN, SVM, and RF classifiers, respectively with a maximum reduced size of the subset being ten only. Comparing the proposed method to eight other feature selection methods, the former achieves better classification accuracy in terms of selecting the most useful but a smaller number of features.

Integrating Sequential Backward Selection (SBS) and CatBoost for Snow Avalanche Susceptibility Mapping at Catchment Scale

Integrating Sequential Backward Selection (SBS) and CatBoost for Snow Avalanche Susceptibility Mapping at Catchment Scale

Electroencephalography microstate alterations reflect potential double-edged cognitive adaptation in Ménière's disease.

To explore the microstate characteristics and underlying brain network activity of Ménière's disease (MD) patients based on high-density electroencephalography (EEG), elucidate the association between microstate dynamics and clinical manifestation, and explore the potential of EEG microstate features as future neurobiomarkers for MD. Thirty-two patients diagnosed with MD and 29 healthy controls (HC) matched for demographic characteristics were included in the study. Dysfunction and subjective symptom severity were assessed by neuropsychological questionnaires, pure tone audiometry, and vestibular function tests. Resting-state EEG recordings were obtained using a 256-channel EEG system, and the electric field topographies were clustered into four dominant microstate classes (A, B, C, and D). The dynamic parameters of each microstate were analyzed and utilized as input for a support vector machine (SVM) classifier to identify significant microstate signatures associated with MD. The clinical significance was further explored through Spearman correlation analysis. MD patients exhibited an increased presence of microstate class C and a decreased frequency of transitions between microstate class A and B, as well as between class A and D. The transitions from microstate class A to C were also elevated. Further analysis revealed a positive correlation between equilibrium scores and the transitions from microstate class A to C under somatosensory challenging conditions. Conversely, transitions between class A and B were negatively correlated with vertigo symptoms. No significant correlations were detected between these characteristics and auditory test results or emotional scores. Utilizing the microstate features identified via sequential backward selection, the linear SVM classifier achieved a sensitivity of 86.21% and a specificity of 90.61% in distinguishing MD patients from HC. We identified several EEG microstate characteristics in MD patients that facilitate postural control yet exacerbate subjective symptoms, and effectively discriminate MD from HC. The microstate features may offer a new approach for optimizing cognitive compensation strategies and exploring potential neurobiological markers in MD.

Interpretable Machine Learning Prediction of Voltage and Specific Capacity for Electrode Materials

AbstractThe application of machine learning (ML) is becoming widespread and playing an important role in the property prediction and design of battery. Average voltage and specific capacity are two important indicators of electrode materials, which determine suitability of electrode materials in batteries. In this study, 4351 data are collected from Material Project and 324 features are extracted from Material Project or calculated by Matminer. Four ML models are discussed for prediction of average voltage and specific capacity. In order to ensure the prediction accuracy and efficiency of ML models, sequential backward selection (SBS) method is introduced to select optimal feature set, which can reduce redundancy features. The combination of Deep Neural Network model and optimal feature set selected by the SBS method achieves accurate prediction and outperforms Crystal Graph Convolutional Neural Network and Graph‐attention Graph Neural Network. The interpretability analysis provides insight into the relationship between features and two target properties. To cope with the poor prediction performance for metal‐ion battery with scarce data, transfer learning is adopted and an excellent improvement is achieved in the prediction in Na‐ion, Mg‐ion and Ca‐ion electrode material. It can be concluded that ML is an effective approach to battery property prediction and design.

Hybrid wavelength selection strategy combined with ATR-FTIR spectroscopy for preliminary exploration of vintage labeling traceability of sauce-flavor baijiu

The allure of substantial profits has perpetuated the illicit trade of counterfeit vintage labels for baijiu. While various approaches have been employed to intelligently ascertain the vintage of baijiu, many of them are both cost-intensive and time-consuming. This work pioneered the use of Attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy, coupled with chemometric analysis, offering a non-destructive and economically viable method for discriminating sauce-flavor baijiu across different aging periods (1-, 2-, and 3-year). In this research, principal component analysis (PCA) was first conducted to explore clustering trends among distinct vintage groups. Subsequently, the effect of spectral pre-processing on modeling performance was explored. For wavelength selection, four wavelength selection methods (ReliefF, random forest variable importance (RFVI), variable importance in projection (VIP), and Venn) were first used to identify the subset of candidate features that potentially best mapped the vintage labels. Immediately following this, to explore the possibility of further improving the identification capabilities of the model as well as to reduce the redundant data that may still be present, sequential backward selection (SBS) was utilized for secondary feature reduction within the subset of candidates. The amalgamation of these two techniques is termed a “hybrid wavelength selection strategy.” Additionally, the dimensionality reduction effects of PCA and kernel principal component analysis (KPCA) were compared to demonstrate the robustness of the proposed method. Finally, classification models such as partial least squares discriminant analysis (PLS-DA), random forest (RF), and grasshopper optimization algorithm-based support vector machine (GOA-SVM) were developed. The results show that the spectral data need not be pre-processed, and the proposed hybrid wavelength selection strategy can further improve the identification ability of the model. Among the many models developed, ReliefF-SBS-GOA-SVM emerged as the most proficient classification model, yielding accuracy, sensitivity, and specificity rates of 94.44%, 95.23%, and 94.44%, respectively. This method not only holds promise for the discrimination of baijiu class attributes such as brand, origin, flavor, and vintage but also exhibits potential applicability in other non-targeted identification studies involving spectroscopy methodologies.

Feature subset selection using heuristic and metaheuristic approaches for diabetes prediction on a binary encoded dataset

The Machine Learning (ML) models are prone to a curse of dimensionality. The dataset with a greater number of features involves more computational cost and it may lead to low performance in the context of prediction accuracy. Therefore, in this research work we have predicted diabetes with more accuracy by using a smaller number of features. The heuristic methods Sequential Forward Selection (SFS), Sequential Backward Selection (SBS) and metaheuristic evolutionary methods — Whale Optimization Algorithm (WOA) and Genetic Algorithm (GA) are used for performing feature subset selection. The Gini index is also used as a filter evaluator. The performance of the feature subsets is analyzed by applying three different types of ML models, Random Forest (RF), Multi-Layer Perceptron (MLP) and K-Nearest Neighbor (KNN). We have predicted type-2 diabetes with an accuracy of 96.82%. Also, we have reduced the number of features up to 67.44% i.e., identified 32.56% most relevant features.

Flexible Ureteroscopy Lithotripsy Operative Time Prediction Model for the Treatment of Kidney Stones

Effective time and resource management is crucial not only in the operating room but also in healthcare supply chains. Healthcare supply chains involve the movement of medical supplies, equipment, and medications from manufacturers to healthcare providers. Effective management is crucial to ensuring that patients receive the care they need promptly. In the operating room, it is essential to have an information process in place to effectively manage time and resources during the current surgical procedure. This paper focuses on developing a predictive model for the operating time of flexible ureteroscopy for kidney stones. The model can forecast surgical and preoperative time based on patient characteristics and surgeon experience. The model can assist in planning ureteroscopy procedures and preventing surgical complications, which is crucial not only for the operating room but also for healthcare supply chains. The paper presents a study that compares different feature selection methods and regression techniques. The study found that sequential backward selection combined with the extra tree regressor was the most effective approach.

Prediction of intraoperative red blood cell transfusion in valve replacement surgery: machine learning algorithm development based on non-anemic cohort.

Our study aimed to develop machine learning algorithms capable of predicting red blood cell (RBC) transfusion during valve replacement surgery based on a preoperative dataset of the non-anemic cohort. A total of 423 patients who underwent valvular replacement surgery from January 2015 to December 2020 were enrolled. A comprehensive database that incorporated demographic characteristics, clinical conditions, and results of preoperative biochemistry tests was used for establishing the models. A range of machine learning algorithms were employed, including decision tree, random forest, extreme gradient boosting (XGBoost), categorical boosting (CatBoost), support vector classifier and logistic regression (LR). Subsequently, the area under the receiver operating characteristic curve (AUC), accuracy, recall, precision, and F1 score were used to determine the predictive capability of the algorithms. Furthermore, we utilized SHapley Additive exPlanation (SHAP) values to explain the optimal prediction model. The enrolled patients were randomly divided into training set and testing set according to the 8:2 ratio. There were 16 important features identified by Sequential Backward Selection for model establishment. The top 5 most influential features in the RF importance matrix plot were hematocrit, hemoglobin, ALT, fibrinogen, and ferritin. The optimal prediction model was CatBoost algorithm, exhibiting the highest AUC (0.752, 95% CI: 0.662-0.780), which also got relatively high F1 score (0.695). The CatBoost algorithm also showed superior performance over the LR model with the AUC (0.666, 95% CI: 0.534-0.697). The SHAP summary plot and the SHAP dependence plot were used to visually illustrate the positive or negative effects of the selected features attributed to the CatBoost model. This study established a series of prediction models to enhance risk assessment of intraoperative RBC transfusion during valve replacement in no-anemic patients. The identified important predictors may provide effective preoperative interventions.

Comparing Performances of Predictive Models of Toxicity after Radiotherapy for Breast Cancer Using Different Machine Learning Approaches.

Different ML models were compared to predict toxicity in RT on a large cohort (n = 1314). The endpoint was RTOG G2/G3 acute toxicity, resulting in 204/1314 patients with the event. The dataset, including 25 clinical, anatomical, and dosimetric features, was split into 984 for training and 330 for internal tests. The dataset was standardized; features with a high p-value at univariate LR and with Spearman ρ>0.8 were excluded; synthesized data of the minority were generated to compensate for class imbalance. Twelve ML methods were considered. Model optimization and sequential backward selection were run to choose the best models with a parsimonious feature number. Finally, feature importance was derived for every model. The model's performance was compared on a training-test dataset over different metrics: the best performance model was LightGBM. Logistic regression with three variables (LR3) selected via bootstrapping showed performances similar to the best-performing models. The AUC of test data is slightly above 0.65 for the best models (highest value: 0.662 with LightGBM). No model performed the best for all metrics: more complex ML models had better performances; however, models with just three features showed performances comparable to the best models using many (n = 13-19) features.

ANFIS-based Framework for the Prediction of Bearing’s Remaining Useful Life

Bearings are critical components extensively used in rotary machines, often being the leading cause of unexpected machine shutdowns. To mitigate system failures, it is crucial to implement effective maintenance strategies. This paper introduces a novel methodology for bearing prognostics, employing Wavelet Packet Decomposition (WPD) for data preprocessing, Sequential Backward Selection (SBS) for feature selection, and Adaptive Neuro-Fuzzy Inference System (ANFIS) networks for prognostic modeling. The proposed approach consists of two key steps. Firstly, the data undergoes preprocessing through Wavelet Packet Decomposition, enhancing the quality and extracting relevant features. Subsequently, the Remaining Useful Life (RUL) of the bearing is predicted using a degradation model. The accuracy of the proposed method is evaluated using a bearing life dataset obtained from a run-to-failure test (IMS dataset). The results demonstrate the remarkable capability of the ANFIS model to learn and accurately estimate the system’s RUL. By leveraging the combined power of WPD, SBS, and ANFIS, this methodology showcases its potential as an effective prognostic tool for bearing health assessment and proactive maintenance planning.

New health indicators for the monitoring of bearing failures under variable loads

Bearings are one of the most critical components in rotating machines. Unexpected failure of this components may cause serious damages and unplanned breakdowns. In this paper, a new method is proposed for bearing fault diagnosis under various loads based on the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and the sequential backward selection (SBS). The CEEMDAN is used for the automatic selection of intrinsic mode functions from vibration signals to build a bank of health indicators. Then, the SBS algorithm is used to select the most relevant indicators of the bearing different failure modes. To test the proposed method accuracy, it was first applied to the Case Western Reserve University dataset. Then, data collected from a dedicated test bench of the Laboratory of Signal and Industrial Process Analysis were introduced to this method to classify different bearing health states. The obtained results show that the proposed method is effective in identifying and classifying bearing faults under various loads with high accuracy. This method can be used for condition monitoring of bearings and prognostics in real industrial applications.

Predicting radiographic outcomes of vertebral body tethering in adolescent idiopathic scoliosis patients using machine learning.

Anterior Vertebral Body Tethering (AVBT) is a growing alternative treatment for adolescent idiopathic scoliosis (AIS), offering an option besides spinal fusion. While AVBT aims to correct spinal deformity through growth correction, its outcomes have been mixed. To improve surgical outcomes, this study aimed to develop a machine learning-based tool to predict short- and midterm spinal curve correction in AIS patients who underwent AVBT surgery, using the most predictive clinical, radiographic, and surgical parameters. After institutional review board approval and based on inclusion criteria, 91 AIS patients who underwent AVBT surgery were selected from the Shriners Hospitals for Children, Philadelphia. For all patients, longitudinal standing (PA or AP, and lateral) and side bending spinal Radiographs were retrospectively obtained at six visits: preop and first standing, one year, two years, five years postop, and at the most recent follow-up. Demographic, radiographic, and surgical features associated with curve correction were collected. The sequential backward feature selection method was used to eliminate correlated features and to provide a rank-ordered list of the most predictive features of the AVBT correction. A Gradient Boosting Regressor (GBR) model was trained and tested using the selected features to predict the final correction of the curve in AIS patients. Eleven most predictive features were identified. The GBR model predicted the final Cobb angle with an average error of 6.3 ± 5.6 degrees. The model also provided a prediction interval, where 84% of the actual values were within the 90% prediction interval. A list of the most predictive features for AVBT curve correction was provided. The GBR model, trained on these features, predicted the final curve magnitude with a clinically acceptable margin of error. This model can be used as a clinical tool to plan AVBT surgical parameters and improve outcomes.

Machine learning-based prediction of effluent total suspended solids in a wastewater treatment plant using different feature selection approaches: A comparative study

Accurately predicting the characteristics of effluent, discharged from wastewater treatment plants (WWTPs) is crucial for reducing sampling requirements, labor, costs, and environmental pollution. Machine learning (ML) techniques can be effective in achieving this goal. To optimize ML-based models, various feature selection (FS) methods are employed. This study aims to investigate the impact of six FS methods (categorized as Wrapper, Filter, and Embedded methods) on the accuracy of three supervised ML algorithms in predicting total suspended solids (TSS) concentration in the effluent of a municipal wastewater treatment plant. Based on the features proposed by each FS method, five distinct scenarios were defined. Within each scenario, three ML algorithms, namely artificial neural network-multi layer perceptron (ANN-MLP), K-nearest neighbors (KNN), and adaptive boosting (AdaBoost) were applied. The features utilized for predicting TSS concentration in the WWTP effluent included BOD5, COD, TSS, TN, NH3 in the influent, and BOD5, COD, residual Cl2, NO3, TN, NH4 in the effluent. To construct the models, the dataset was randomly divided into training and testing subsets, and K-fold cross-validation was employed to control overfitting and underfitting. The evaluation metrics that are used are root mean squared error (RMSE), mean absolute error (MAE), and correlation coefficient (R2). The most efficient scenario was identified as Scenario IV, with the Sequential Backward Selection FS method. The features selected by this method were CODe, BOD5e, BOD5i, TNi. Furthermore, the ANN-MLP algorithm demonstrated the best performance, achieving the highest R2 value. This algorithm exhibited acceptable performance in both the training and testing subsets (R2 = 0.78 and R2 = 0.8, respectively).

Accurate Mental Stress Detection Using Sequential Backward Selection and Adaptive Synthetic Methods.

The daily experience of mental stress profoundly influences our health and work performance while concurrently triggering alterations in brain electrical activity. Electroencephalogram (EEG) is a widely adopted method for assessing cognitive and affective states. This study delves into the EEG correlates of stress and the potential use of resting EEG in evaluating stress levels. Over 13 weeks, our longitudinal study focuses on the real-life experiences of college students, collecting data from each of the 18 participants across multiple days in classroom settings. To tackle the complexity arising from the multitude of EEG features and the imbalance in data samples across stress levels, we use the sequential backward selection (SBS) method for feature selection and the adaptive synthetic (ADASYN) sampling algorithm for imbalanced data. Our findings unveil that delta and theta features account for approximately 50% of the selected features through the SBS process. In leave-one-out (LOO) cross-validation, the combination of band power and pair-wise coherence (COH) achieves a maximum balanced accuracy of 94.8% in stress-level detection for the above daily stress dataset. Notably, using ADASYN and borderline synthesized minority over-sampling technique (borderline-SMOTE) methods enhances model accuracy compared to the traditional SMOTE approach. These results provide valuable insights into using EEG signals for assessing stress levels in real-life scenarios, shedding light on potential strategies for managing stress more effectively.

Efficient prediction of SOC and aggregate OC components by continuous wavelet transform spectra under different feature selection methods

Soil Organic Carbon (SOC) fixation plays a vital role in reducing greenhouse gas emissions and the formation of soil aggregates is recognized as a crucial process in SOC fixation. In the current study, we characterized spectral information on SOC and aggregate Organic carbon (OC) components by using a rapid and robust method. Soil samples were collected from different layers to obtain the SOC and aggregate OC components to record corresponding spectra. The Continuous Wavelet Transform (CWT) spectra were then combined with various characteristic wavelength selection methods to build Multivariate Linear Regression (MLR), BP neural network (BP), and Random Forest (RF) models. Finally, the optimal combination of characteristic wavelengths based on the best-performing model for each index was determined. The results demonstrated that the model's accuracy, which was built using the set of characteristic wavelengths of the best model for all metrics, was higher than that of the best model for a single metric. This was due to the correlation between SOC and aggregate OC components. The best set of characteristic wavelengths were those selected by the Sequential Backward Selection method with wavelet decomposition scales of 23, 26, and 27, and those selected by the Minimum Redundancy Maximum Relevance method with wavelet decomposition scale 27. The RF model for Free silt and clay particles organic carbon (FSP-OC) showed the highest accuracy among the four SOC prediction models. In addition, it was discovered that as the wavelet decomposition scales become more similar, the extracted characteristic wavelengths also become more similar, resulting in similar model predictions. In summary, the accuracy of the predictive model for SOC and its constituents can be demonstrably improved by reasonable combination of characteristic wavelengths selection method and wavelet decomposition scale.

Fitting a collider in a quantum computer: tackling the challenges of quantum machine learning for big datasets.

Current quantum systems have significant limitations affecting the processing of large datasets with high dimensionality, typical of high energy physics. In the present paper, feature and data prototype selection techniques were studied to tackle this challenge. A grid search was performed and quantum machine learning models were trained and benchmarked against classical shallow machine learning methods, trained both in the reduced and the complete datasets. The performance of the quantum algorithms was found to be comparable to the classical ones, even when using large datasets. Sequential Backward Selection and Principal Component Analysis techniques were used for feature's selection and while the former can produce the better quantum machine learning models in specific cases, it is more unstable. Additionally, we show that such variability in the results is caused by the use of discrete variables, highlighting the suitability of Principal Component analysis transformed data for quantum machine learning applications in the high energy physics context.

Prediction of malaria positivity using patients’ demographic and environmental features and clinical symptoms to complement parasitological confirmation before treatment

BackgroundCurrent malaria diagnosis methods that rely on microscopy and Histidine Rich Protein-2 (HRP2)-based rapid diagnostic tests (RDT) have drawbacks that necessitate the development of improved and complementary malaria diagnostic methods to overcome some or all these limitations. Consequently, the addition of automated detection and classification of malaria using laboratory methods can provide patients with more accurate and faster diagnosis. Therefore, this study used a machine-learning model to predict Plasmodium falciparum (Pf) antigen positivity (presence of malaria) based on sociodemographic behaviour, environment, and clinical features.MethodData from 200 Nigerian patients were used to develop predictive models using nested cross-validation and sequential backward feature selection (SBFS), with 80% of the dataset randomly selected for training and optimisation and the remaining 20% for testing the models. Outcomes were classified as Pf-positive or Pf-negative, corresponding to the presence or absence of malaria, respectively.ResultsAmong the three machine learning models examined, the penalised logistic regression model had the best area under the receiver operating characteristic curve for the training set (AUC = 84%; 95% confidence interval [CI]: 75–93%) and test set (AUC = 83%; 95% CI: 63–100%). Increased odds of malaria were associated with higher body weight (adjusted odds ratio (AOR) = 4.50, 95% CI: 2.27 to 8.01, p < 0.0001). Even though the association between the odds of having malaria and body temperature was not significant, patients with high body temperature had higher odds of testing positive for the Pf antigen than those who did not have high body temperature (AOR = 1.40, 95% CI: 0.99 to 1.91, p = 0.068). In addition, patients who had bushes in their surroundings (AOR = 2.60, 95% CI: 1.30 to 4.66, p = 0.006) or experienced fever (AOR = 2.10, 95% CI: 0.88 to 4.24, p = 0.099), headache (AOR = 2.07; 95% CI: 0.95 to 3.95, p = 0.068), muscle pain (AOR = 1.49; 95% CI: 0.66 to 3.39, p = 0.333), and vomiting (AOR = 2.32; 95% CI: 0.85 to 6.82, p = 0.097) were more likely to experience malaria. In contrast, decreased odds of malaria were associated with age (AOR = 0.62, 95% CI: 0.41 to 0.90, p = 0.012) and BMI (AOR = 0.47, 95% CI: 0.26 to 0.80, p = 0.006).ConclusionNewly developed routinely collected baseline sociodemographic, environmental, and clinical features to predict Pf antigen positivity may be a valuable tool for clinical decision-making.

Investigating the feasibility of differentiating MS active lesions from inactive ones using texture analysis and machine learning methods in DWI images

BackgroundMagnetic resonance imaging (MRI) is commonly used in conjunction with a gadolinium-based contrast agent (GBCA) to distinguish active multiple sclerosis (MS) lesions. However, recent studies have raised concerns regarding the long-term effects of the accumulation of GBCA in the body. Thus, the purpose of this study is to investigate the possibility of using texture analysis in diffusion-weighted imaging (DWI) and machine learning algorithms to discriminate active from inactive MS lesions without the use of GBCA. MethodsTo achieve this purpose, we introduce an image processing pipeline. In the proposed pipeline, following registration and alignment of slices, MS lesions from DWI images are segmented and quantized. Next, different texture analysis methods are employed to extract texture features from the lesions. Then, a two-stage feature reduction method is applied, in which the first stage involves a statistical t-test and the second stage relies on principal component analysis (PCA), sequential forward selection (SFS), sequential backward selection (SBS), and ReliefF algorithms. Finally, we use five classifiers logistic regression (LR), support vector machine (SVM), decision tree (DT), K nearest neighbor (KNN), and linear discriminant analysis (LDA) in a 5-fold cross-validation procedure to determine active and inactive MS lesions. ResultsIn this study, we collected and prepared 255 active/inactive MS lesions from MRI scans of 34 patients diagnosed with MS, with a mean age of 35.56±10.89. Among 89 texture features extracted, 63 features showed statistically significant differences between the means of active and inactive lesions (P<0.05). The SVM classifier with the PCA feature reduction algorithm demonstrated the best performance with an average accuracy of 0.960 (±0.024), specificity and precision of 1.0, sensitivity of 0.913 (±0.053), and AUC of 0.957 (±0.027). ConclusionOur study indicates that DWI changes detected using texture analysis-based machine learning models can precisely differentiate active from inactive MS lesions. This finding provides valuable clinical information for the early diagnosis and effective monitoring of MS disease.

An interpretable machine learning model for predicting bond strength of CFRP-steel epoxy-bonded interface

This study develops an interpretable machine learning model for predicting the bond strength of CFRP-steel epoxy bonding interfaces and reveals key bond parameters. A total of 302 sets of experimental data were collected from the existing literature, including 16 influencing factors (i.e. the “features”) for the bond strength of the CFRP-steel bonded interface. Firstly, the sequential backward selection algorithms based on random forest (RF), maximum information coefficient (MIC), and distance coefficient (DC) were assessed. Then, Catboost and RF-based models were optimized and evaluated to determine the best prediction model. Finally, Shapley Additive Explanation (SHAP) was used to interpret the Catboost-based model. The results showed that the average Coefficient of Determination (R2) of the prediction for the test set by the Catboost and RF-based models was 95.5 % and 91.5 %, respectively, indicating the Catboost-based model has a higher prediction accuracy. The SHAP analysis indicates that the bond length, bond width, and the elastic modulus of the adhesive are the critical bonding parameters governing the bond strength of CFRP-steel epoxy-bonded interfaces. The Pareto diagram shows that the effective bond length is about 120 mm. This study provides valuable guidance for the design of bond performance and process optimization of CFRP-steel epoxy bonding interfaces.

Sequential backward feature selection for optimizing permanent strain model of unbound aggregates

This study proposes a novel framework for identifying the optimal feature set required to predict the permanent strain of unbound aggregates. An experimental database consisting of 16 input features is preprocessed and the performance of 10 machine learning models is evaluated. The best-performing model is then paired with a sequential backward selection algorithm to determine the optimal feature set for predicting the permanent strain. Finally, the selected features are used to predict the permanent strain, and the performance is compared with those obtained from the principal components analysis. Six features are selected as the optimal feature set. Furthermore, the selected features accurately predict permanent strain with a root mean square error value of 0.014, which is smaller than those obtained from principal components analysis. Thus, the feature selection approach for machine learning models effectively predicts the permanent strain of unbound aggregates using a limited set of input features.