Developing a novel approach for missing data imputation of solar radiation: A hybrid differential evolution algorithm based eXtreme gradient boosting model

Machine learning‐based streamflow forecasting using CMIP6 scenarios: Assessing performance and improving hydrological projections and climate change

“…Konya Closed Basin is located between latitudes 36°51' N and 39°29' N, longitudes 31°36′ and 34°52′ E, in the Central Anatolia Region of Turkey. Sakarya and Kızılırmak surround the basin from the north, Eastern Mediterranean from the south, Kızılırmak and Seyhan from the east, Antalya and Akarçay Basins from the west (Başakın et al, 2023). Konya, Isparta, Niğde, Ankara, Aksaray, Nevşehir and Karaman provinces are in the basin.…”

“…The second‐order Taylor expansion is applied by XGBoost to get closer to the loss function value. XGBoost allows cross‐validation to be performed to get the error of the validation set by minimizing the over‐fitting of the model in each iteration within the boosting process (Başakın et al, 2023; Serencam et al, 2022).…”

“…The gradient Boosting model is a commonly used tree-based ML method based on gradient boosting logic to achieve higher precision approximations for setting an optimal prediction model (Bas ¸akın et al, 2023;Chen & Guestrin, 2016). The computational efficiency of XGBoost is its dominant feature relative to other boosting algorithms such as gradient boosting, adaptive boosting, and stochastic gradient In this study, the Gradient Boosting model was meticulously constructed using the powerful scikit-learn library, a widely used and publicly available tool for machine learning in Orange tool.…”