What is it about?
This research focuses on predicting actual evapotranspiration using readily available meteorological variables. After considering their correlation with the target variable and addressing multicollinearity, a select few variables are chosen. Additionally, a systematic literature review identifies six machine learning techniques that are experimented with in this study.
Featured Image
Photo by Maksym Kaharlytskyi on Unsplash
Why is it important?
The results of our study reveal the vital meteorological variables for predicting actual evapotranspiration. The automatic systematic literature review outcomes align with our selection of candidate variables and identify suitable machine learning methods. We experiment with six-time series-based machine learning methods and compare their performance.
Perspectives
This article provides valuable assistance to researchers engaged in machine learning applications for climate data, specifically focusing on multivariate time series challenges. The research findings hold substantial significance for policymakers, irrigation managers, and water resource managers, enabling them to make informed decisions regarding the efficient utilization of water resources for agricultural purposes, ultimately enhancing productivity levels and reducing the severity of the drought.
Mr. CHALACHEW MULUKEN LIYEW
University of Torino (UNITO)
Read the Original
This page is a summary of: Multivariate Time Series Evapotranspiration Forecasting using Machine Learning Techniques, March 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3555776.3577838.
You can read the full text:
Resources
Contributors
The following have contributed to this page
