Identification of explanatory variables for DMU preparation process evaluation by using machine learning techniques

Abstract

Being able to estimate a priori the impact of DMU preparation scenarios for a dedicated activity would help identifying the best scenario from the beginning. Machine learning techniques are a means to a priori evaluate a DMU preparation process without to perform it by predicting its criteria of evaluation. For that, a representative database of examples must be developed that contains the right explanative and output variables. However, the key explanative variables are not clearly identified. This paper proposes a method for the selection of the most significant explanatory variables among all the database variables. In addition to using these variables for learning, this will allow to formalize the knowledge.