Affiliations: [a] Electrical Engineering Department, German Jordanian University, Amman, Jordan | [b] Faculty of Computer Sciences and Informatics, Amman Arab University, Amman, Jordan | [c] School of Computer Sciences, Universiti Sains Malaysia, Pulau Pinang, Malaysia | [d] Information Technology Department, Amman Arab University, Amman, Jordan | [e] Computer Engineering Department, The University of Jordan, Amman, Jordan
Correspondence:
[*]
Corresponding author. Khalid A. Darabkh, Computer Engineering Department, The University of Jordan, Amman, Jordan. E-mail: [email protected].
Abstract: Weather prediction is paramount for many applications and scenarios, among them is agriculture. In order to efficiently irrigate the crops with the exact needed water amount, weather forecasting can be used to optimize the quantity of required irrigation water such that the crops are neither dried up nor over-irrigated. This paper proposes a Machine Learning (ML)-based weather forecasting model, which utilizes the Social Spider Algorithm-Least Square-Support Vector Machine (SSA-LS-SVM) algorithm. The simulation results are used to predict the prime weather and soil parameters such as the atmospheric temperature, pressure, and soil humidity for 24, 48, and 72 hours based on previous 39 days’ hourly data for Amman city. The predicted values showed low relative mean square errors compared with the actual values and the LS-SVM predictor.
Keywords: Weather forecasting, prediction, smart irrigation, artificial intelligence, social spider algorithm-least square support vector machine, least square support vector machine
