El modelo de up-lifting aplicado a un score de riesgo

Abstract

Data mining has been widely used to optimize customer management in order to maximize return on investment. In particular, this paper deals with the use of scoring models in product marketing campaigns. Models are typically developed to identify the characteristics of customers who are most likely to incur an event (default, purchase a product, withdraw a product, etc.). While these models are useful for identifying the customers to be targeted by a marketing campaign, this campaign can be targeted to customers who have already decided what action to take with respect to the event in question (in this case, purchase of a product), regardless of whether or not they receive the campaign (e.g., email, call). We propose the application of a methodology to identify customers whose decisions will be positively influenced by campaigns. The proposed methodology is simple to implement and can be used in combination with the most commonly used supervised learning algorithms. This methodology can provide the telecommunications sector with a simple but significant methodological improvement to optimize its marketing actions.