A Decision Tree-Based Data Mining Approach to Estimating Dental Tourists’ Satisfaction State

Abstract

This study introduces a data mining approach to estimate dental tourists’ satisfaction. In detail, the study combines statistical techniques and a decision tree model to ensure a complete estimation pipeline. The study was conducted using survey feedback from a total 314 dental tourists, who visited a private dental clinic in Kuşadası, in the Aydın province of Turkey. The collected data were analysed by using Pearson coefficient, Kolmogorov-Smirnov test, Mann-Whitney U test, and Kruskal-Wallis test. Next, factor analysis was used to identify various factors that established the input parameters of the decision tree model. In terms of statistical findings, it was found that there was no significant relationship between age and the satisfaction of the dental tourists. It was also found that there was a significant difference in satisfaction by gender, and dental tourists’ satisfaction levels differed significantly based on citizenship, education level, and income status as these differences varied across relevant demographic classes. At the data mining application stage, a total of four factors determined through factor analysis were used to train a decision tree model and the model achieved respectively 96%, and 80% accuracy rates for during training and testing data. The study concludes that various demographic data and factors influence dental tourists’ service satisfaction, and data mining can help in effectively estimating satisfaction levels for potential dental tourists.