Master’s Thesis at the University of Basrah Explores the Use of Artificial Intelligence for Regression Model Estimation
A master’s thesis at the College of Administration and Economics, Department of Statistics, University of Basrah, investigated the use of artificial intelligence techniques to estimate certain nonparametric regression models through practical application.
The thesis, submitted by student Dmoo Mohammad Radi, aimed to evaluate several machine learning models within the framework of artificial intelligence to estimate and analyze earthquake and seismic data.
The study compared four nonparametric models: the Multivariate Adaptive Regression Splines (MARS) model, Kernel Regression, the K-Nearest Neighbors (KNN) algorithm, and Support Vector Regression (SVR), using various statistical performance evaluation indicators.
The results demonstrated that the MARS model outperformed the other models across all indicators, achieving the highest coefficient of determination and the lowest error values. The study concluded that the MARS model is the most suitable for handling complex data sets.
Department of Media and Government Communication
