Variable selection and/or importance measurement of input variables to a machine learning model has become the focus of much research due to the European regulation on privacy protection. Having a good model is no longer enough, one must also explain its decisions. Therefore, there are many intelligibility algorithms available today. Among these, there are many algorithms for estimating Shapley values, a method of intelligibility based on the theory of cooperative games. This article proposes a comparison of Shapley values (in the particular case of the Naive Bayes Classifier) with a frequently used indicator: the weight of evidence.