Investigation of estimation performance for different soil areas

[ X ]

Tarih

2020

Dergi Başlığı

Dergi ISSN

Cilt Başlığı

Yayıncı

Springer

Erişim Hakkı

info:eu-repo/semantics/closedAccess

Özet

Soil plays a vital role in the climate system. This paper performs decision tree regression to estimate soil moisture (SM) by considering different parameters that include air temperature, time, relative humidity, and soil temperature. Besides, this paper investigates the effects of the parameters of decision tree regression by utilizing the response surface. The obtained estimation results of two distinct soil areas, Field and Forest, indicate that two different soil areas have distinct estimation quality. Furthermore, numerical results of the training stage show that the estimation of SM for Field and Forest soil performing decision tree regression offers 0.0019 and 0.0025 mean absolute error (MAE), respectively. Moreover, numerical results show that the interaction of the parameters of the performed algorithm plays a vital role in the estimation stage of Field and Forest soils.

Açıklama

Anahtar Kelimeler

Artificial learning, Decision tree regression, Estimation, Performance evaluation, Soil moisture

Kaynak

Environmental Monitoring And Assessment

WoS Q Değeri

Q3

Scopus Q Değeri

Q2

Cilt

192

Sayı

5

Künye