Integration of machine learning algorithms as an innovative tool for teaching marketing in business administration programs

Abstract

The present research study aims to use machine learning algorithms in the teaching of marketing in business administration programs to predict customer behavior. The methodology applied is CRISP-DM (Cross Industry Standard Process for Data Mining), which is used to guide data mining. As a results, four machine learning algorithms are obtained: k-nearest neighbors (KNN), decision trees, random forests, and a voting regressor. These algorithms are used to predict how much a customer will spend while identifying her/his preferences and predicting her/his behavior. The assessment metric used is the mean square error (MSE). The results show that the best performing algorithm is random forest. In conclusion, incorporating machine learning in the teaching of marketing improves student analytical abilities while providing valuable tools for making strategic decisions.