Title: Location method of garden air pollution source based on gradient lifting regression tree algorithm

Authors: Xiang Huang; Liuzhen Li

Addresses: Department of Horticulture, Xinyang Agriculture and Forestry University, Xinyang 464000, China ' Xuchang Forestry and Flower Gardening Development Centre, Xuchang 461000, China

Abstract: In order to improve the location accuracy of garden air pollution sources, this paper proposes a new method of garden air pollution source location based on gradient lifting regression tree algorithm. Firstly, the infrared remote sensing spectrometer is used to collect the air pollution data in the garden. Secondly, the decision tree algorithm is used to construct the regression tree of garden pollution data. The least square regression tree is used as the base learner, and the gradient lifting regression tree is constructed by iterative calculation. Finally, the gradient value of the gradient lifting regression tree is calculated and fitted with the regression tree to obtain the node location of the regression tree, which is the location of the pollution source. The experimental results show that, compared with the traditional location methods, the pollution source location accuracy of this method is higher, which is always maintained at more than 96%.

Keywords: gradient lifting regression tree algorithm; garden air environment; pollution source location; gradient value.

DOI: 10.1504/IJETM.2023.134326

International Journal of Environmental Technology and Management, 2023 Vol.26 No.6, pp.445 - 456

Received: 27 May 2022
Accepted: 03 Oct 2022
Published online: 18 Oct 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article