Feature selection for global tropospheric ozone prediction based on the BO-XGBoost-RFE algorithm

Zhang, Biao; Zhang, Ying; Jiang, Xuchu

doi:10.1038/s41598-022-13498-2

Cited by 23 publications

(10 citation statements)

References 21 publications

(18 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, it often outperforms other ML algorithms, which is why it has won several Kaggle competitions. However, it has a few limitations, including its high number of hyperparameters, making it difficult to tune [86], [87].…”

Section: Procedurementioning

confidence: 99%

A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects

2022

View full text Add to dashboard Cite

Ensemble learning techniques have achieved state-of-the-art performance in diverse machine learning applications by combining the predictions from two or more base models. This paper presents a concise overview of ensemble learning, covering the three main ensemble methods: bagging, boosting, and stacking, their early development to the recent state-of-the-art algorithms. The study focuses on the widely used ensemble algorithms, including random forest, adaptive boosting (AdaBoost), gradient boosting, extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost). An attempt is made to concisely cover their mathematical and algorithmic representations, which is lacking in the existing literature and would be beneficial to machine learning researchers and practitioners.

show abstract

Section: Procedurementioning

confidence: 99%

A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects

2022

View full text Add to dashboard Cite

show abstract

“…Due to the intrinsic properties of eXtreme Gradient Boosting (XGBoost), feature selection in this research was based on the importance scores of the features. This technique has been widely used in various applications and has demonstrated satisfactory performance (Chen et al 2020a, Dai et al 2022, Zhang et al 2022. Consequently, the XGBoost model was used to fit the sample data.…”

Section: Feature Selectionmentioning

confidence: 99%

Exploring influential factors of CO2 emissions in China’s cities using machine learning techniques

xiang

et al. 2023

Preprint

View full text Add to dashboard Cite

Investigating the factors that exert an influence on CO2 emissions represents a critical undertaking for the formulation of effective policies aimed at reducing such emissions. Numerous past studies have attempted to explore the potential relationships between CO2 emissions and a variety of potential influencing factors. However, many of these investigations have been constrained by the inability to fit large sample datasets, as well as by the limitations of conventional research methods in addressing non-linear relationships. The aforementioned insufficiencies have resulted in a situation wherein a restricted set of factors can be examined in tandem within a singular model. This restriction has impeded the comprehensive investigation of multiple variables and their potential interrelationships. To bridge these research gaps, this research employed machine learning models to fit voluminous datasets concerning urban CO2 emissions and socioeconomic factors. This research also incorporated explainable techniques to disentangle the intricate relationships between the factors under investigation. The results demonstrated that urban fiscal structure, urban energy consumption, urban land use, and urban capital accumulation represented the primary factors driving urban CO2 emissions. The ALE test was employed to discern the specific relationships between these factors and CO2 emissions, ultimately revealing that almost all of these factors positively impacted CO2 emissions as their values rise. In both univariate and bivariate ALE tests, Expenditures on Science and Technology had emerged as a highly influential feature in terms of its impact on urban CO2 emissions. While the interaction effects elicited by Expenditures on Science and Technology had the potential to curb urban CO2 emissions in several ways, this research identified that these effects may be constrained. The relationship identified between urban socio-economic development and CO2 emissions also indicated that achieving low-carbon development in urban areas necessitates optimal resource allocation and the transformation of energy consumption structures. Furthermore, each city needs to tailor its low-carbon development pathway to its distinct characteristics.

show abstract

“…The Bayesian optimization framework is used to establish and update the probabilistic surrogate model based on previous evaluations of the objective function [ 47 ], and to actively select the evaluation points with the most global “potential” through the acquisition function. Bayesian optimization can effectively use prior information to judge the uncertainty of the unknown region and obtain the optimal solution within a few evaluations.…”

Section: Thermal Conductivity Identification Based On An Optimization...mentioning

confidence: 99%

Improved Bayesian Optimization Framework for Inverse Thermal Conductivity Based on Transient Plane Source Method

Liu

Lyu

et al. 2023

Entropy

View full text Add to dashboard Cite

In order to reduce the errors caused by the idealization of the conventional analytical model in the transient planar source (TPS) method, a finite element model that more closely represents the actual heat transfer process was constructed. The average error of the established model was controlled at below 1%, which was a significantly better result than for the analytical model, which had an average error of about 5%. Based on probabilistic optimization and heuristic optimization algorithms, an optimization model of the inverse heat transfer problem with partial thermal conductivity differential equation constraints was constructed. A Bayesian optimization algorithm with an adaptive initial population (BOAAIP) was proposed by analyzing the influencing factors of the Bayesian optimization algorithm upon inversion. The improved Bayesian optimization algorithm is not affected by the range and individuals of the initial population, and thus has better adaptability and stability. To further verify its superiority, the Bayesian optimization algorithm was compared with the genetic algorithm. The results show that the inversion accuracy of the two algorithms is around 3% when the thermal conductivity of the material is below 100 Wm−1K−1, and the calculation speed of the improved Bayesian optimization algorithm is three to four times faster than that of the genetic algorithm.

show abstract

Feature selection for global tropospheric ozone prediction based on the BO-XGBoost-RFE algorithm

Cited by 23 publications

References 21 publications

A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects

A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects

Exploring influential factors of CO2 emissions in China’s cities using machine learning techniques

Improved Bayesian Optimization Framework for Inverse Thermal Conductivity Based on Transient Plane Source Method

Contact Info

Product

Resources

About