This paper shows how an AI machine can be used as an action-oriented rule finder tool to support managerial decision-making and scenario depiction with an application in the context of loyalty programs in retail chains. In a market oriented perspective this paper suggests how to exploit knowledge from data using a parsimonious model that provides explanations in a human-like language. The model is based on analogies between human cognitive processes and boosting heuristics coming from Artificial Intelligence. Real data from grocery channels are used to test the model against generalized linear models, showing good performances and giving recommendations coherent with managerial practice.
A rule finder machine to exploit market knowledge in the grocery retail context
VISENTIN, MARCO
2005
Abstract
This paper shows how an AI machine can be used as an action-oriented rule finder tool to support managerial decision-making and scenario depiction with an application in the context of loyalty programs in retail chains. In a market oriented perspective this paper suggests how to exploit knowledge from data using a parsimonious model that provides explanations in a human-like language. The model is based on analogies between human cognitive processes and boosting heuristics coming from Artificial Intelligence. Real data from grocery channels are used to test the model against generalized linear models, showing good performances and giving recommendations coherent with managerial practice.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.