Narjes Nabipour

TL;DR: The current research proposes the state-of-the-art ensemble models of boosted generalized linear model (GLMBoost) and random forest (RF) and Bayesian generalizedlinear model (BayesGLM) methods for higher performance modeling and a pre-processing method is used to eliminate redundant variables from the modeling process.

TL;DR: Three indices of drought are modeled using Support Vector Regression, Gene Expression Programming, and M5 model trees and the results indicate that SPI delivered higher accuracy than SSI.

TL;DR: Novel integrative flood susceptibility prediction models based on multi-time resampling approaches, random subsampling and bootstrapping algorithms, integrated with machine learning models outperformed the benchmark models such as Standalone GAM, MARS, BRT, multilayer perceptron (MLP) and support vector machine (SVM).

TL;DR: It was concluded that the WOA optimization algorithm could improve the prediction accuracy of the MLP model and may be recommended for accurate wind speed prediction.

TL;DR: A groundwater compatibility study was conducted by analyzing Electrical conductivity (EC), total dissolved solids (TDS), Chloride (Cl), Calcium (Ca), Magnesium (Mg), Sodium (Na), Potassium (K), Sulfate (SO4), Total hardness (TH), Bicarbonate (HCO3), pH, carbonate (CO3) and Sodium Adsorption Ratio (SAR) obtained from 39 wells in the time period from 2003 to 2014 as discussed by the authors.