Analysis of X-ray diffraction (XRD) data using artificial intelligence techniques

This thesis investigates the analysis of X-Ray Diffraction (XRD) data using artificial neural networks (ANN) and k-Nearest Neighbors (kNN) algorithms. While XRD is widely used to determine crystal structures, analyzing complex and nonlinear data remains a challenge with traditional methods. In this study, machine learning techniques based on artificial intelligence are integrated into the analysis of XRD data to overcome these limitations. XRD data obtained from volcanosedimentary rock samples collected from the Erzurum region were analyzed using ANN and kNN algorithms. ANN demonstrated high accuracy in predicting material properties and effectively modeled complex relationships between crystal structures. The kNN algorithm, on the other hand, was successfully employed in classifying phases and crystal structures within the XRD data. Experiments with different K values revealed that smaller K values were more sensitive to local structures, while larger K values provided more generalizable predictions. Advanced cross-validation techniques were used to evaluate model performance, reducing the risk of overfitting and improving the generalization capability of the models. The ANN and kNN models applied in this study achieved significant success in detecting phase transitions and classifying crystal structures. This thesis demonstrates the applicability of artificial intelligence-based methods in the analysis of XRD data and contributes to more in-depth and accurate outcomes in scientific research within this field.