Boosting-Based Machine Learning Applications in Polymer Science: A Review

Polymers, Journal Year: 2025, Volume and Issue: 17(4), P. 499 - 499

Published: Feb. 14, 2025

The increasing complexity of polymer systems in both experimental and computational studies has led to an expanding interest machine learning (ML) methods aid data analysis, material design, predictive modeling. Among the various ML approaches, boosting methods, including AdaBoost, Gradient Boosting, XGBoost, CatBoost LightGBM, have emerged as powerful tools for tackling high-dimensional complex problems science. This paper provides overview applications science, highlighting their contributions areas such structure-property relationships, synthesis, performance prediction, characterization. By examining recent case on techniques this review aims highlight potential advancing characterization, optimization materials.

Language: Английский

---

Predicting and Mapping of Soil Organic Matter with Machine Learning in the Black Soil Region of the Southern Northeast Plain of China

Agronomy, Journal Year: 2025, Volume and Issue: 15(3), P. 533 - 533

Published: Feb. 22, 2025

The estimation of soil organic matter (SOM) content is essential for understanding the chemical, physical, and biological functions soil. It also an important attribute reflecting quality black In this study, machine learning algorithms support vector (SVM), neural network (NN), decision tree (DT), random forest (RF), extreme gradient boosting (GBM), generalized linear model (GLM) were used to study accurate prediction SOM in Tieling County, City, Liaoning Province, China. models trained by using 1554 surface samples 19 auxiliary variables. Recursive feature elimination was as a selection method identify effective results showed that Normalized Difference Vegetation Index (NDVI) elevation key Based on 10-fold cross-validation, RF had highest accuracy. terms accuracy, coefficient determination 0.77, root mean square error 2.85. average 20.15 g/kg. spatial distribution shows higher concentrated east west, while lower found middle. cultivated land than land.

Language: Английский

---

Novel Biomarker Prediction for Lung Cancer Using Random Forest Classifiers

Cancer Informatics, Journal Year: 2023, Volume and Issue: 22

Published: Jan. 1, 2023

Lung cancer is considered the most common and deadliest type. could be mainly of 2 types: small cell lung non-small cancer. Non-small affected by about 85% while only 14%. Over last decade, functional genomics has arisen as a revolutionary tool for studying genetics uncovering changes in gene expression. RNA-Seq been applied to investigate rare novel transcripts that aid discovering genetic occur tumours due different cancers. Although helps understand characterise expression involved diagnostics, biomarkers remains challenge. Usage classification models uncover classify based on levels over The current research concentrates computing transcript statistics from files with normalised fold change genes identifying quantifiable differences between reference genome samples. collected data analysed, machine learning were developed causing NSCLC, SCLC, both or neither. An exploratory analysis was performed identify probability distribution principal features. Due limited number features available, all them used predicting class. To address imbalance dataset, an under-sampling algorithm Near Miss carried out dataset. For classification, primarily focused 4 supervised algorithms: Logistic Regression, KNN classifier, SVM classifier Random Forest additionally, ensemble algorithms considered: XGboost AdaBoost. Out these, weighted metrics considered, showing 87% accuracy best performing thus predict NSCLC SCLC. dataset restrict any further improvement model's precision. In our present study using values (LogFC, P Value) feature sets Classifier BRAF, KRAS, NRAS, EGFR predicted possible ATF6, ATF3, PGDFA, PGDFD, PGDFC PIP5K1C SCLC transcriptome analysis. It gave precision 91.3% 91% recall after fine tuning. Some CDK4, CDK6, BAK1, CDKN1A, DDB2.

Language: Английский

---

Predicting adhesion strength of micropatterned surfaces using gradient boosting models and explainable artificial intelligence visualizations

Materials Today Communications, Journal Year: 2023, Volume and Issue: 36, P. 106545 - 106545

Published: June 28, 2023

Fibrillar dry adhesives are widely used due to their effectiveness in air and vacuum conditions. However, performance depends on various factors. Previous studies have proposed analytical methods predict adhesion strength micro-patterned surfaces. the method lacks interpretation which parameters critical. This research utilizes gradient-boosting machine learning (ML) algorithms accurately strength. Additionally, explainable (XML) employed interpret underlying reasoning behind predictions. The analysis demonstrates that gradient boosting models achieve a high correlation coefficient (R > 0.95) predicting pull-off force use of XML provides insights into importance features, interactions, contributions specific novel, explainable, data-driven approach holds potential for real-time applications, aiding identification critical features govern fibrillar adhesives. Furthermore, it improves end-users' confidence by offering human-comprehensible explanations facilitates understanding among non-technical audiences.

Language: Английский

---

Greedy-AutoML: A novel greedy-based stacking ensemble learning framework for assessing soil liquefaction potential

Engineering Applications of Artificial Intelligence, Journal Year: 2022, Volume and Issue: 119, P. 105732 - 105732

Published: Dec. 21, 2022

Language: Английский

---

Novel diabetes classification approach based on CNN-LSTM: Enhanced performance and accuracy

Diagnostyka, Journal Year: 2024, Volume and Issue: 25(1), P. 1 - 15

Published: Feb. 11, 2024

1. Diabetes Statistics. Center for Diabetic Empowerment Education. https://ceed-diabete.org/fr/le.... Google Scholar

Language: Английский

---

Cone penetration test-based assessment of liquefaction potential using machine and hybrid learning approaches

Multiscale and Multidisciplinary Modeling Experiments and Design, Journal Year: 2024, Volume and Issue: 7(4), P. 3841 - 3864

Published: April 26, 2024

Language: Английский

---

Three-dimensional undrained stability analysis of circular tunnel heading in anisotropic and heterogeneous clay: FELA, ANN, MARS, and XGBoost

Modeling Earth Systems and Environment, Journal Year: 2024, Volume and Issue: 10(4), P. 5503 - 5527

Published: July 2, 2024

Language: Английский

---

Improving Aggregate Abrasion Resistance Prediction via Micro-Deval Test Using Ensemble Machine Learning Techniques

Engineering Journal, Journal Year: 2024, Volume and Issue: 28(3), P. 15 - 24

Published: March 1, 2024

Aggregate is the most extracted material from world's mines and widely used in civil construction projects.The Micro-Deval abrasion test (MD) one of important tests that provides characteristics crushed aggregates show their resistance against mechanical abrasive factors such as repeated impact loading.The various on properties has led researchers to seek correlations, often focusing limited data samples, leading reduced accuracy.This study employs machine learning (ML) methods predict MD values, considering diverse aggregate properties.Various ensemble ML were applied, revealing exceptional performance stacking model, which achieved an R 2 score 0.95 predicting resistance.The feature importance analysis highlights influence Magnesium Sulfate Soundness (MSS), Water Absorption (ABS), Los Angeles Abrasion (LAA) suggesting use multiple could yield a more dependable assessment durability.

Language: Английский

---

Machine Learning Techniques for Soil Characterization Using Cone Penetration Test Data

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(14), P. 8286 - 8286

Published: July 18, 2023

Seismic response assessment requires reliable information about subsurface conditions, including soil shear wave velocity (Vs). To properly assess seismic response, engineers need accurate Vs, an essential parameter for evaluating the propagation of waves. However, measuring Vs is generally challenging due to complex and time-consuming nature field laboratory tests. This study aims predict using machine learning (ML) algorithms from cone penetration test (CPT) data. The utilized four ML algorithms, namely Random Forests (RFs), Support Vector Machine (SVM), Decision Trees (DT), eXtreme Gradient Boosting (XGBoost), Vs. These models were trained on 70% datasets, while their efficiency generalization ability assessed remaining 30%. hyperparameters each model fine-tuned through Bayesian optimization with k-fold cross-validation techniques. performance was evaluated eight different metrics, root mean squared error (RMSE), absolute (MAE), percentage (MAPE), coefficient determination (R2), index (PI), scatter (SI), A10−I, U95. results demonstrated that RF consistently performed well across all metrics. It achieved high accuracy lowest level errors, indicating superior precision in predicting SVM XGBoost also exhibited strong performance, slightly higher metrics compared model. DT poorly, rates uncertainty Based these results, we can conclude highly effective at accurately CPT data minimal input features.

Language: Английский

---