Proceedings of 2nd International Multi-Disciplinary Conference Theme: Integrated Sciences and Technologies, IMDC-IST 2021, 7-9 September 2021, Sakarya, Turkey

Research Article

Precipitation Forecast for Thi-Qar Province of Iraq Utilizing Machine Learning Approaches

Download338 downloads
  • @INPROCEEDINGS{10.4108/eai.7-9-2021.2314897,
        author={Anwar Alnawas and Nassir Jabir Al-Khafaji and Hayder Hussein Azeez},
        title={Precipitation Forecast for Thi-Qar Province of Iraq Utilizing Machine Learning Approaches},
        proceedings={Proceedings of 2nd International Multi-Disciplinary Conference Theme: Integrated Sciences and Technologies, IMDC-IST 2021, 7-9 September 2021, Sakarya, Turkey},
        publisher={EAI},
        proceedings_a={IMDC-IST},
        year={2022},
        month={1},
        keywords={machine learning data mining rapidminer rainfall rate thi-qar},
        doi={10.4108/eai.7-9-2021.2314897}
    }
    
  • Anwar Alnawas
    Nassir Jabir Al-Khafaji
    Hayder Hussein Azeez
    Year: 2022
    Precipitation Forecast for Thi-Qar Province of Iraq Utilizing Machine Learning Approaches
    IMDC-IST
    EAI
    DOI: 10.4108/eai.7-9-2021.2314897
Anwar Alnawas1,*, Nassir Jabir Al-Khafaji1, Hayder Hussein Azeez1
  • 1: Nassriyah Technical Institute, Southern Technical University, Iraq
*Contact email: anwar.alnawas@stu.edu.iq

Abstract

Rainfall is considered a main to provide water in rivers along with Iraqi territory. The unpredictable amount of rainfall due to climate change can cause either overflow or dry in the rivers. Although, there are a lot of electronic devices that have harnessed the prediction of precipitation using weather conditions such as humidity pressure, and temperature. Regrettably, these classical methods cannot work efficiently, so exploiting machine learning techniques can predict accurate outcomes. Therefore, predictions of data-based models using deep learning algorithms are promising for these purposes. This empirical study seeks to build a precipitation prediction model using a deep learning mechanism through utilizing historical weather data. Deep learning outperformed other classifiers based on the findings collected. The current study's experiment yielded accurate findings of up to 91.59% when testing the model with actual weather data within the specified period.