Application of Machine Learning Methods to Rainfall Prediction: The Case of Iraq
Abstract
Forecasting rainfall averages obtained from weather variables using machine learning methods at a potential location is required to gain deeper insight into one of the important weather ingredients. This study proposed an approach to predict and assess rainfall values using the daily dataset (dew point and relative humidity) at three locations in Iraq, namely Baghdad, Najaf, and Wasit. Linear and nonlinear machine learning approaches established based on linear regression were compared with the KNN method. According to the statistical indicators, the KNN models were slightly superior to linear models, with a weakness in prediction performance. In addition, both proposed methods are useful for extracting annual rainfall averages.
Fields of Study & Subjects
Hydrological Forecasting Using AI
Data Mining and Machine Learning Applications
Precipitation Measurement and Analysis
Publication details
- Journal / source
- International Journal of Applied Science and Research
- Volume
- 09
- Issue
- 01
- Pages
- 141–145
- Publication date
- 1 January 2026
- Language
- EN