Rainfall Classification Using Machine Learning Algorithms on Data Mining Platforms
IEEE Canadian Journal of Electrical and Computer Engineering, cilt.48, sa.2, ss.109-114, 2025 (Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 48 Sayı: 2
- Basım Tarihi: 2025
- Doi Numarası: 10.1109/icjece.2025.3558882
- Dergi Adı: IEEE Canadian Journal of Electrical and Computer Engineering
- Derginin Tarandığı İndeksler: Scopus
- Sayfa Sayıları: ss.109-114
- Anahtar Kelimeler: Classification, machine learning algorithms, open-source data mining software, weather forecast
- İstanbul Üniversitesi-Cerrahpaşa Adresli: Evet
Özet
Weather conditions directly affect sectors such as agriculture and transport. With climate change, unpredictability is increasing and traditional calculation methods may not be sufficient. In addition to some statistical methods, machine learning algorithms are also used for weather forecasting. This study attempts to classify precipitation using machine learning algorithms on selected meteorological data. The models used are K-nearest neighbors (KNNs), support vector machine (SVM), and multilayer perceptron (MLP). These models were implemented on four different open-source and free data mining platforms. These platforms are Altair AI Studio (formerly Rapidminer), Knime, Orange, and Weka. The dataset includes parameters such as pressure, temperature, humidity, number of rainy days, cloudiness rate, and year and month information. According to the values of these parameters, the data were classified as less rainy, rainy, and very rainy.