Implementation of data mining methods on household labor research data and comparison of models

In today' s conditions, increasing the need for such as assessing, analyzing and making forward estimates of data, has increased the demand for data mining methods. In this study, classification and regression estimating methods of data mining are discussed, C5.0 decision tree, CHAID decision tree, Logistic Regression and Bayesian Networks methods are examined and the modeling successes were compared by applying them on a real data set. Additionally, data on the Household Labor Force Survey compiled and implemented by Turkish Statistical Institute for 2014, 2015 and 2016 years. Also, within the scope of this study to determine the labor situation in Turkey, both the employed labor force status by making regarding the classification of individuals outside employment profile is intended to be an accurate estimate of the labor situation as a result of unknown individual models. When the study results were evaluated, it was calculated that the model obtained as a result of the C5.0 decision tree method had the most successful classification prediction even though the model successes were relatively close to each other in comparison with various criteria.