Conventional Machine Learning Models Research Articles

Employing Xception convolutional neural network through high-precision MRI analysis for brain tumor diagnosis

The classification of brain tumors from medical imaging is pivotal for accurate medical diagnosis but remains challenging due to the intricate morphologies of tumors and the precision required. Existing methodologies, including manual MRI evaluations and computer-assisted systems, primarily utilize conventional machine learning and pre-trained deep learning models. These systems often suffer from overfitting due to modest medical imaging datasets and exhibit limited generalizability on unseen data, alongside substantial computational demands that hinder real-time application. To enhance diagnostic accuracy and reliability, this research introduces an advanced model utilizing the Xception architecture, enriched with additional batch normalization and dropout layers to mitigate overfitting. This model is further refined by leveraging large-scale data through transfer learning and employing a customized dense layer setup tailored to effectively distinguish between meningioma, glioma, and pituitary tumor categories. This hybrid method not only capitalizes on the strengths of pre-trained network features but also adapts specific training to a targeted dataset, thereby improving the generalization capacity of the model across different imaging conditions. Demonstrating an important improvement in diagnostic performance, the proposed model achieves a classification accuracy of 98.039% on the test dataset, with precision and recall rates above 96% for all categories. These results underscore the possibility of the model as a reliable diagnostic tool in clinical settings, significantly surpassing existing diagnostic protocols for brain tumors.

Feature Transfer Learning for Fatigue Life Prediction of Additive Manufactured Metals With Small Samples

ABSTRACTA feature transfer learning (FTL)‐based model is proposed to address small‐sample problems in fatigue life prediction of additively manufactured (AM) metals. Transfer component analysis (TCA) is studied for data alignment before model training. Correspondingly, two TCA improvement strategies are further considered to aggregate training data from distinct AM processing conditions. An experimental database consisting of 103 fatigue data is built for model evaluation. The results demonstrate that the proposed model outperforms conventional machine learning models and other transfer learning‐based models in terms of accuracy and data demand, showing good applicability for AM fatigue life assessment.

MST-m6A: A Novel Multi-Scale Transformer-based Framework for Accurate Prediction of m6A Modification Sites Across Diverse Cellular Contexts

N6-methyladenosine (m6A) modification, a prevalent epigenetic mark in eukaryotic cells, is crucial in regulating gene expression and RNA metabolism. Accurately identifying m6A modification sites is essential for understanding their functions within biological processes and the intricate mechanisms that regulate them. Recent advances in high-throughput sequencing technologies have enabled the generation of extensive datasets characterizing m6A modification sites at single-nucleotide resolution, leading to the development of computational methods for identifying m6A RNA modification sites. However, most current methods focus on specific cell lines, limiting their generalizability and practical application across diverse biological contexts. To address the limitation, we propose MST-m6A, a novel approach for identifying m6A modification sites with higher accuracy across various cell lines and tissues. MST-m6A utilizes a multi-scale transformer-based architecture, employing dual k-mer tokenization to capture rich feature representations and global contextual information from RNA sequences at multiple levels of granularity. These representations are then effectively combined using a channel fusion mechanism and further processed by a convolutional neural network to enhance prediction accuracy. Rigorous validation demonstrates that MST-m6A significantly outperforms conventional machine learning models, deep learning models, and state-of-the-art predictors. We anticipate that the high precision and cross-cell-type adaptability of MST-m6A will provide valuable insights into m6A biology and facilitate advancements in related fields. The proposed approach is available at https://github.com/cbbl-skku-org/MST-m6A/ for prediction and reproducibility purposes.

Fake News Detection Using Machine Learning

Abstract: The identification of fake news has grown in importance as false information circulates quickly via digital channels. The goal of this project is to combine deep learning and machine learning to create an automated system that can identify news stories as legitimate or fraudulent. The preprocessed text data in the dataset, which comprises of tagged news items, was done using methods such tokenization, stop word removal, and padding. A Convolutional Neural Network (Conv1D), which is based on deep learning, was combined with more conventional machine learning models, such as Naive Bayes, in a hybrid approach. After converting the text input into a series of tokens, semantic meanings were captured by using word embeddings. Important patterns from the word sequences were intended to be captured by the Conv1D model with global max pooling, whereas the Naive Bayes also used in the project.

Rapid estimation of DON content in wheat flour using close‐range hyperspectral imaging and machine learning

AbstractFusarium head blight (FHB) is one of the most destructive fungal diseases affecting wheat (Triticum aestivum). Moreover, it is notorious for producing mycotoxin deoxynivalenol (DON), posing a significant global threat to food and feed safety. Traditional methods like enzyme‐linked immunosorbent assay (ELISA) and gas chromatography‐mass spectrometry (GC‐MS) are commonly used to assess DON levels in grain or flour samples and are time‐consuming and expensive. Therefore, a faster, cost‐effective method to estimate DON content is needed, especially for enhancing breeding efforts to reduce DON levels in wheat. In this study, we envisioned integrating close‐range hyperspectral imaging with deep learning (DL) models to estimate DON content in wheat meal/flour. We selected 243 advanced breeding lines from the South Dakota State University (SDSU) wheat breeding program that were evaluated in FHB nurseries (2019–2020 and 2020–2021). The wheat meal samples were analyzed for DON content using GC‐MS and subsequently subjected to close‐range hyperspectral imaging. We evaluated three conventional machine learning (ML), two DL models and data augmentation. Among the conventional ML models, partial least squares regression (PLSR) (with R2P = 0.88 and 0.90 for original and augmented datasets, respectively) demonstrated the highest prediction accuracies for DON content. However, the one‐dimensional convolutional neural network (1D‐CNN) achieved the highest prediction accuracies (R2P = 0.90 and = 0.96 for original and augmented datasets, respectively) compared to all tested models and demonstrated the lowest error. In conclusion, integration of advanced hyperspectral imaging with ML approaches exhibits significant potential for high‐throughput and cost‐effective estimation of DON content in wheat, thereby accelerating wheat breeding efforts for reduced DON levels.

Forecasting Patient Early Readmission from Irish Hospital Discharge Records Using Conventional Machine Learning Models.

Predicting patient readmission is an important task for healthcare risk management, as it can help prevent adverse events, reduce costs, and improve patient outcomes. In this paper, we compare various conventional machine learning models and deep learning models on a multimodal dataset of electronic discharge records from an Irish acute hospital. We evaluate the effectiveness of several widely used machine learning models that leverage patient demographics, historical hospitalization records, and clinical diagnosis codes to forecast future clinical risks. Our work focuses on addressing two key challenges in the medical fields, data imbalance and the variety of data types, in order to boost the performance of machine learning algorithms. Furthermore, we also employ SHapley Additive Explanations (SHAP) value visualization to interpret the model predictions and identify both the key data features and disease codes associated with readmission risks, identifying a specific set of diagnosis codes that are significant predictors of readmission within 30 days. Through extensive benchmarking and the application of a variety of feature engineering techniques, we successfully improved the area under the curve (AUROC) score from 0.628 to 0.7 across our models on the test dataset. We also revealed that specific diagnoses, including cancer, COPD, and certain social factors, are significant predictors of 30-day readmission risk. Conversely, bacterial carrier status appeared to have minimal impact due to lower case frequencies. Our study demonstrates how we effectively utilize routinely collected hospital data to forecast patient readmission through the use of conventional machine learning while applying explainable AI techniques to explore the correlation between data features and patient readmission rate.

Artificial intelligence for predicting arctic permafrost and active layer temperatures along the Alaskan North Slope

AbstractClimate change in Arctic regions poses a threat to permafrost stability, potentially leading to issues such as infrastructure damage, altered ecosystems, and increased greenhouse gas emissions. The challenge of monitoring permafrost temperatures and identifying critical environmental factors is accentuated by the scarcity of subsurface thermal data in remote Arctic locales. To alleviate this lack of soil thermal data we set out to combine in situ observed soil temperatures with MERRA-2 reanalysis data and Machine Learning (ML) to develop local and season-specific subsurface soil temperature predictions in Alaska. First, reanalysis features are selected based on surface energy budget physics. Coupled with Julian calendar dates, reanalysis features are preprocessed and then fed to the ML prediction model. We conducted a series of computational experiments and tested the results against performance benchmarks to determine the optimal prediction models, length of look-back period of model inputs, and the training set size. Six conventional ML models (e.g., GBDT, RF, SVR) and five statistical baselines (e.g. ARIMA) using in situ time series of soil temperature data ranging from 0 - 1 m depth are considered for the prediction task. The models are trained using in situ soil temperatures at depths between 0 and 1 m, spanning 16+ years, from field sites at Deadhorse and Toolik Lake, Alaska. All prediction performances are assessed using root mean squared error (RMSE) and mean absolute error (MAE). Results show that locally trained ML models can estimate shallow soil temperatures with an average error of $$RMSE=1.308^{\circ }$$ R M S E = 1 . 308 ∘ C.

Comparison of deep and conventional machine learning models for prediction of one supply chain management distribution cost

Strategic supply chain management (SCM) is essential for organizations striving to optimize performance and attain their goals. Prediction of supply chain management distribution cost (SCMDC) is one branch of SCM and it’s essential for organizations striving to optimize performance and attain their goals. For this purpose, four machine learning algorithms, including random forest (RF), support vector machine (SVM), multilayer perceptron (MLP) and decision tree (DT), along with deep learning using convolutional neural network (CNN), was used to predict and analyze SCMDC. A comprehensive dataset consisting of 180,519 open-source data points was used for analyze and make the structure of each algorithm. Evaluation based on Root Mean Square Error (RMSE) and Correlation coefficient (R2) show the CNN model has high accuracy in SCMDC prediction than other models. The CNN algorithm demonstrated exceptional accuracy on the test dataset, with an RMSE of RMSE of 0.528 and an R2 value of 0.953. Notable advantages of CNNs include automatic learning of hierarchical features, proficiency in capturing spatial and temporal patterns, computational efficiency, robustness to data variations, minimal preprocessing requirements, end-to-end training capability, scalability, and widespread adoption supported by extensive research. These attributes position the CNN algorithm as the preferred choice for precise and reliable SCMDC predictions, especially in scenarios requiring rapid responses and limited computational resources.

Natural language processing‐based deep transfer learning model across diverse tabular datasets for bond strength prediction of composite bars in concrete

AbstractAs conventional machine learning models often struggle with scarcity and structural variation of training data, this paper proposes a novel regression transfer learning framework called transferable tabular regressor (TransTabRegressor) to address this challenge. The TransTabRegressor integrates natural language processing (NLP) for feature encoding, transformer for enhanced feature representation, and deep learning (DL) for robust modeling, facilitating effective transfer learning across tabular datasets using reducing input parameters. By leveraging the NLP data processor, the framework embeds both parameter names and values, enabling it to recognize and adapt to different expressions of similar parameters. For instance, the bond strength of fiber‐reinforced polymer (FRP) bars embedded in ultra‐high‐performance concrete (UHPC) is critical for ensuring the integrity of FRP‐UHPC structures. While pullout tests are widely adopted for their simplicity to generate substantial data, beam tests provide a closer approximation to actual stress conditions but are more complex thus resulting in limited data size. As a verification, the framework is applied to predict the bond strength of FRP bars embedded in UHPC using limited beam test data. A pre‐trained model is first established using 479 pieces of pullout test data. Subsequently, two transfer learning models are developed by fine‐tuning on 115 pieces of beam test data, where 66 correspond to concrete splitting failure and 49 correspond to pullout failure. For comparative analysis, XGBoost and neural network models are directly trained on the beam test data. Evaluation results demonstrate that the transfer learning models achieve significantly improved prediction accuracy and generalization capability. This study significantly highlights the effectiveness of the proposed TransTabRegressor in handling data scarcity and variability in input parameters across various engineering applications.

GAN-Augmented Naïve Bayes for identifying high-risk coronary artery disease patients using CT angiography data

Coronary artery disease (CAD) is one of the most common cardiovascular disorders affecting millions of individuals globally. It is the leading cause of mortality in both the wealthy and impoverished nations. CAD patients exhibit a wide range of symptoms, some of which are not evident until a major incident occurs. The development of techniques for early detection and precise diagnosis is heavily dependent on research. The proposed system introduces a novel approach, Generative Adversarial Networks Augmented Naïve Bayes (GAN-ANB), to classify high-risk CAD patients using Coronary Computed Tomography Angiography (CCTA) imaging data. The database included images from Coronary Computed Tomography Angiography (CCTA) records of 5,000 individuals. The developed GAN framework consists of a generator to generate synthetic patient profiles, and a discriminator to distinguish between genuine and synthetic profiles to improve the identification of high-risk CAD patients. Adding synthetic data to the training process allowed the discriminator to be utilized further to improve predictive modeling. The performance of the GAN-enhanced prediction model was assessed using accuracy, sensitivity, specificity, and area under the Receiver Operating Characteristic curve (ROC). The model exhibited an outstanding Dice Similarity Coefficient (0.91), Mean Intersection Over Union (0.90), recall (0.96), and precision (0.98) in differentiating between high-risk and low-risk individuals. The identification of high-risk patients with CAD is greatly enhanced by the integration of GANs with clinical and imaging data. ROC of 0.99 was achieved by the GAN-ANB model, which outperformed conventional machine learning models, was achieved using the GAN-ANB model. High cholesterol level, diabetes, and some CCTA-derived imaging characteristics, including plaque load and luminal stenosis, were among the major predictors. This method offers a powerful tool for early diagnosis and intervention, potentially leading to improved patient outcomes and lower healthcare expenditure.

Convolutional Neural Networks for Real Time Classification of Beehive Acoustic Patterns on Constrained Devices.

Recent research has demonstrated the effectiveness of convolutional neural networks (CNN) in assessing the health status of bee colonies by classifying acoustic patterns. However, developing a monitoring system using CNNs compared to conventional machine learning models can result in higher computation costs, greater energy demand, and longer inference times. This study examines the potential of CNN architectures in developing a monitoring system based on constrained hardware. The experimentation involved testing ten CNN architectures from the PyTorch and Torchvision libraries on single-board computers: an Nvidia Jetson Nano (NJN), a Raspberry Pi 5 (RPi5), and an Orange Pi 5 (OPi5). The CNN architectures were trained using four datasets containing spectrograms of acoustic samples of different durations (30, 10, 5, or 1 s) to analyze their impact on performance. The hyperparameter search was conducted using the Optuna framework, and the CNN models were validated using k-fold cross-validation. The inference time and power consumption were measured to compare the performance of the CNN models and the SBCs. The aim is to provide a basis for developing a monitoring system for precision applications in apiculture based on constrained devices and CNNs.

Strategies for designing machine learning models in renewable energy with insufficient data

In the area of renewable energy including wind, solar, and electricity, Machine Learning has established a far-reaching conceptual and algorithmic design framework. However, the design of Machine Learning architectures typically relies on substantial amounts of data. The data acquisition process often involves complex tasks of gathering, cleaning, and labeling data. This challenge is exacerbated in the energy sector due to concerns related to data security and the uniqueness of development initiatives, leading to a potential scarcity of available data. Quite commonly, researchers encounter a vast array of innovative projects that have not been studied in the past and therefore only limited data become available. Therefore, there is an urgent and highly motivated need to develop machine learning strategies that can construct reliable models in the presence of limited data. In this study, three machine learning models tailored for small-scale data scenarios are developed and analyzed. The first model, a fuzzy cluster ensemble approach, focuses on utilizing small training samples treated as the core of clustering. By applying clustering algorithms, one derives membership functions reflecting the structure of data that facilitate accurate predictions. The second model, a kernel integration framework, combines kernel functions with limited training data to calculate activation levels, enabling accurate prediction outputs. For both models, the final predictions are generated by applying weighted aggregations, where the weights are derived from the respective membership grades and activation levels of the data. Additionally, a neural network contrastive learning model is introduced, which enhances the training process by employing pairwise comparisons among data points. This method effectively increases the amount of useable training data, improving the model's ability to make accurate predictions. Together, these approaches provide a robust solution for making reliable predictions even in data-scarce scenarios within renewable energy research. Moreover, the predictive performance of these models is quantified using the Root Mean Square Error criterion. This study also assesses the performance of the developed models by forming a granular characterization of results delivered by the designed models. A comprehensive suite of experimental studies is included, accompanied by an exhaustive comparative analysis. Compared to conventional machine learning models, especially neural networks, the designed models offer reliable results even with limited training data, showing an improvement of around 20 %, thus laying the groundwork for digitization processes in the realm of sustainable energy.

Efficacy of automated machine learning models and feature engineering for diagnosis of equivocal appendicitis using clinical and computed tomography findings

This study evaluates the diagnostic efficacy of automated machine learning (AutoGluon) with automated feature engineering and selection (autofeat), focusing on clinical manifestations, and a model integrating both clinical manifestations and CT findings in adult patients with ambiguous computed tomography (CT) results for acute appendicitis (AA). This evaluation was compared with conventional single machine learning models such as logistic regression(LR) and established scoring systems such as the Adult Appendicitis Score(AAS) to address the gap in diagnostic approaches for uncertain AA cases. In this retrospective analysis of 303 adult patients with indeterminate CT findings, the cohort was divided into appendicitis (n = 115) and non-appendicitis (n = 188) groups. AutoGluon and autofeat were used for AA prediction. The AutoGluon-clinical model relied solely on clinical data, whereas the AutoGluon-clinical-CT model included both clinical and CT data. The area under the receiver operating characteristic curve (AUROC) and other metrics for the test dataset, namely accuracy, sensitivity, specificity, PPV, NPV, and F1 score, were used to compare AutoGluon models with single machine learning models and the AAS. The single ML models in this study were LR, LASSO regression, ridge regression, support vector machine, decision tree, random forest, and extreme gradient boosting. Feature importance values were extracted using the “feature_importance” attribute from AutoGluon. The AutoGluon-clinical model demonstrated an AUROC of 0.785 (95% CI 0.691–0.890), and the ridge regression model with only clinical data revealed an AUROC of 0.755 (95% CI 0.649–0.861). The AutoGluon-clinical-CT model (AUROC 0.886 with 95% CI 0.820–0.951) performed better than the ridge model using clinical and CT data (AUROC 0.852 with 95% CI 0.774–0.930, p = 0.029). A new feature, exp(-(duration from pain to CT)3 + rebound tenderness), was identified (importance = 0.049, p = 0.001). AutoML (AutoGluon) and autoFE (autofeat) enhanced the diagnosis of uncertain AA cases, particularly when combining CT and clinical findings. This study suggests the potential of integrating AutoML and autoFE in clinical settings to improve diagnostic strategies and patient outcomes and make more efficient use of healthcare resources. Moreover, this research supports further exploration of machine learning in diagnostic processes.

Enhanced gastric cancer classification and quantification interpretable framework using digital histopathology images

Recent developments have highlighted the critical role that computer-aided diagnosis (CAD) systems play in analyzing whole-slide digital histopathology images for detecting gastric cancer (GC). We present a novel framework for gastric histology classification and segmentation (GHCS) that offers modest yet meaningful improvements over existing CAD models for GC classification and segmentation. Our methodology achieves marginal improvements over conventional deep learning (DL) and machine learning (ML) models by adaptively focusing on pertinent characteristics of images. This contributes significantly to our study, highlighting that the proposed model, which performs well on normalized images, is robust in certain respects, particularly in handling variability and generalizing to different datasets. We anticipate that this robustness will lead to better results across various datasets. An expectation-maximizing Naïve Bayes classifier that uses an updated Gaussian Mixture Model is at the heart of the suggested GHCS framework. The effectiveness of our classifier is demonstrated by experimental validation on two publicly available datasets, which produced exceptional classification accuracies of 98.87% and 97.28% on validation sets and 98.47% and 97.31% on test sets. Our framework shows a slight but consistent improvement over previously existing techniques in gastric histopathology image classification tasks, as demonstrated by comparative analysis. This may be attributed to its ability to capture critical features of gastric histopathology images better. Furthermore, using an improved Fuzzy c-means method, our study produces good results in GC histopathology picture segmentation, outperforming state-of-the-art segmentation models with a Dice coefficient of 65.21% and a Jaccard index of 60.24%. The model’s interpretability is complemented by Grad-CAM visualizations, which help understand the decision-making process and increase the model’s trustworthiness for end-users, especially clinicians.

Comparative Analysis of Conventional Machine Learning and Graph Neural Network Models for Perovskite Property Prediction

Comparative Analysis of Conventional Machine Learning and Graph Neural Network Models for Perovskite Property Prediction

David vs. Goliath: comparing conventional machine learning and a large language model for assessing students' concept use in a physics problem.

Large language models have been shown to excel in many different tasks across disciplines and research sites. They provide novel opportunities to enhance educational research and instruction in different ways such as assessment. However, these methods have also been shown to have fundamental limitations. These relate, among others, to hallucinating knowledge, explainability of model decisions, and resource expenditure. As such, more conventional machine learning algorithms might be more convenient for specific research problems because they allow researchers more control over their research. Yet, the circumstances in which either conventional machine learning or large language models are preferable choices are not well understood. This study seeks to answer the question to what extent either conventional machine learning algorithms or a recently advanced large language model performs better in assessing students' concept use in a physics problem-solving task. We found that conventional machine learning algorithms in combination outperformed the large language model. Model decisions were then analyzed via closer examination of the models' classifications. We conclude that in specific contexts, conventional machine learning can supplement large language models, especially when labeled data is available.

Vertex-based graph neural network classification model considering structural topological features for structural optimization

Traditional surrogate models always face the challenge of low accuracy when dealing with high-dimensional problems in structural optimization, this study aims to overcome this problem and proposes a vertex-based graph neural network (GNN) classification model. In contrast to conventional machine learning models that treat design variables as independent inputs, the proposed model develops a vertex-based graph representation to transform structural topological features and critical physical information into the graph data. According to a message passing mechanism based on the graph convolutional, it can extract the correlations among design variables and enhance its capability in handling high-dimensional structural optimization problems. Three truss examples, including a 10-bar with 10 variables, a 600-bar with 25 variables, and a 942-bar with 59 variables, are utilized to investigate the performance of the proposed surrogate model. The results demonstrate that the GNN-based surrogate model outperforms traditional machine learning approaches, particularly in the two high-dimensional problems, showcasing its superior ability to capture complex variable correlations and handle high-dimensional structural optimization tasks. Moreover, the proposed method significantly reduces the computational expenses by over 60% compared to conventional metaheuristic algorithms, while yielding optimal designs with comparable quality. These results demonstrate the efficiency and effectiveness of the GNN-based surrogate model in tackling complex, high-dimensional structural optimization problems.

Improved Prediction for Nonlinear Thermal History of Mg Alloy Subjected to Multiple Electropulses

The repeated rapid heating and nonlinear cooling associated with electropulsing treatment (EPT) hinder the accurate modeling of temperature variations based on a constitutive approach using physical models. This objective is not attainable through conventional machine learning (ML) approaches as well, such as artificial neural network (ANN) and long short‐term memory (LSTM). The present study investigates the intricate nonlinear thermal history of Mg alloys induced by EPT to improve a predictive model under multipulse conditions. Furthermore, it enhances the predictivity and generalizability of a given problem using an ML model called i‐LSTM by specifically tailoring a dataset. The i‐LSTM model performance surpasses that of the constitutive approaches and conventional ML models when predicting the complicated thermal history of EPT specimens in a wide variety of processing parameters. The mechanisms underlying this improvement are discussed based on several experimental datasets. This study not only addresses challenges in existing models but also offers a promising avenue for understanding and predicting the thermal dynamics of Mg alloys subjected to the multipulse EPT process.

Credit card default prediction using ML and DL techniques

The banking sector is widely acknowledged for its intrinsic unpredictability and susceptibility to risk. Bank loans have emerged as one of the most recent services offered over the past several decades. Banks typically serve as intermediaries for loans, investments, short-term loans, and other types of credit. The usage of credit cards is experiencing a steady increase, thereby leading to a rise in the default rate that banks encounter. Although there has been much research investigating the efficacy of conventional Machine Learning (ML) models, there has been relatively less emphasis on Deep Learning (DL) techniques. The application of DL approaches to credit card default prediction has not been extensively researched despite their considerable potential in numerous fields. Moreover, the current literature frequently lacks particular information regarding the DL structures, hyperparameters, and optimization techniques employed. To predict credit card default, this study evaluates the efficacy of a DL model and compares it to other ML models, such as Decision Tree (DT) and Adaboost. The objective of this research is to identify the specific DL parameters that contribute to the observed enhancements in the accuracy of credit card default prediction. This research makes use of the UCI ML repository to access the credit card defaulted customer dataset. Subsequently, various techniques are employed to preprocess the unprocessed data and visually present the outcomes through the use of exploratory data analysis (EDA). Furthermore, the algorithms are hypertuned to evaluate the enhancement in prediction. We used standard evaluation metrics to evaluate all the models. The evaluation indicates that the AdaBoost and DT exhibit the highest accuracy rate of 82 ​% in predicting credit card default, surpassing the accuracy of the ANN model, which is 78 ​%.

Evaluating the strength of subgrade soil for pavement design: an analysis using support vector regression model optimized by bayesian algorithm

Determining the strength of subgrade soil, crucial for flexible pavement design, traditionally relies on time-consuming and labour-intensive California Bearing Ratio (CBR) tests. The limitations of linear and non-linear regression-based prediction methods have led to an increased importance of soft computing techniques in providing flexible and consistent CBR predictions for practical engineering. This paper proposes a hybrid model that combines the Bayesian Optimization algorithm (BOA) with support vector regression (SVR) to predict CBR for subgrade soil in pavement design. A database of 238 soil test datasets from northern Algeria was utilised to develop the BOA-SVR model. Using the BOA technique to optimize hyperparameters can improve the model’s performance. The developed hybrid model closely matched experimental results, with a coefficient of determination of 89% and a mean squared error of 5.77, indicating high predictive accuracy. Comparative analysis shows that the BOA-SVR model performs better than conventional and other machine learning models in predicting CBR in tested subgrade soils, in terms of accuracy and robustness. Finally, the optimal subset of features from the initial dataset combined with the BOA-SVR model is determined using an exhaustive feature selection method that includes cross-validation. The results of the experiment indicate that the accuracy of the hybrid model that has been developed is improved by a subset of only seven features from the twelve initial features, this allows for the reduction of computing time and storage space.