Robust Spatiotemporal Estimation of PM Concentrations Using Boosting-Based Ensemble Models

Abstract

Particulate matter (PM) as an air pollutant is harmful to the human body as well as to the ecosystem. It is crucial to understand the spatiotemporal PM distribution in order to effectively implement reduction methods. However, ground-based air quality monitoring sites are limited in providing reliable concentration values owing to their patchy distribution. Here, we aimed to predict daily PM10 concentrations using boosting algorithms such as gradient boosting machine (GBM), extreme gradient boost (XGB), and light gradient boosting machine (LightGBM). The three models performed well in estimating the spatial contrasts and temporal variability in daily PM10 concentrations. In particular, the LightGBM model outperformed the GBM and XGM models, with an adjusted R2 of 0.84, a root mean squared error of 12.108 μg/m2, a mean absolute error of 8.543 μg/m2, and a mean absolute percentage error of 16%. Despite having high performance, the LightGBM model showed low spatial prediction accuracy near the southwest part of the study area. Additionally, temporal differences were found between the observed and predicted values at high concentrations. These outcomes indicate that such methods can provide intuitive and reliable PM10 concentration values for the management, prevention, and mitigation of air pollution. In the future, performance accuracy could be improved through consideration of different variables related to spatial and seasonal characteristics.

Depression assessment using integrated multi-featured EEG bands deep neural network models: Leveraging ensemble learning techniques

“…The model with high accuracy is given higher weight to optimize the overall model. The ACC (Accuracy) weighted method is an alternative approach, akin to the MAE weighted method The ACC weighted method is learned from the SWE (Self-Weighted Ensemble) method [36] which is based on the F1-score to calculate the performance. The aim of the ACC weighted method is to reduce the weight of inaccurate learners and boost the reasonably-accurate weak learners so as to increase the recognition rate of the overall model.…”

Distinguishing the main climatic drivers to the variability of gross primary productivity at global FLUXNET sites

“…For example, once there are two features with 99% of correlation, decision tree will choose only one of them as statistic feature within the processes of deciding upon a split. Recently, the XGB method was used to address problems in atmospheric environment (Park et al 2021) and terrestrial ecosystems (Wang et al 2022b) and showed better potentials than traditional methods. Here, we apply the XGB model to reproduce site-level GPP based on eight climatic drivers including TA, diffuse radiation (SWdif), direct radiation (SWdir), VPD, CO 2 , WS, air pressure (P) and PRE.…”

Data Diversity in Convolutional Neural Network Based Ensemble Model for Diabetic Retinopathy

“…It incrementally constructs an ensemble by iteratively training a new model to emphasize misclassified training samples from previous models. Boosting helps reduce bias (underfitting) in classification problems [ 31 ].…”