Defining customer profile with artificial intelligence and data mining techniques: Customer relations management dynamic modeli̇ng application

The evolution of the Information Technology Service Management approach is that Artificial Intelligence-supported models are taking an increasingly effective role in management processes. In thesis study, it is important to study the basic techniques of this transformation while analyzing the effects of the decrease in the services provided by human experts and the increase in the impact of artificial intelligence-based decision support systems, especially in the management of customer support tickets. Various regression and classification methods are experimentally analyzed for predicting the completion time of customer support tickets, determining the priority and assigning the tickets to the most appropriate expert. The results are compared with the Coefficient of Determination, Accuracy, Precision and F1 Score criteria and a system architecture using Decision Tree Regression, Artificial Neural Network Classification Algorithm and a BERT based pre-trained large language model is proposed.