Analyzing ranking data using decision tree

Abstract

Ranking/preference data arises from many applications in marketing, psychology and politics. We establish a new decision tree model for the analysis of ranking data by adopting the concept of classification and regression tree [2]. We modify the existing splitting criteria, Gini and entropy, which can precisely measure the impurity of a set of ranking data. Two types of impurity measures for ranking data are introduced, namely n-wise and top-k measures. Minimal cost-complexity pruning is used to find the optimum-sized tree. In model assessment, the area under the ROC curve (AUC) is applied to evaluate the tree performance. The proposed methodology is implemented to analyze a partial ranking dataset of Inglehart's items collected in the 1993 International Social Science Programme survey. Change in importance of item values with country, age and level of education are identified.