Título: Comparing Mutual Information and Delta Test for Variable Selection to Determine Client Confidence in e-Bank
Autores: Liebana, F.; Guillen, A.; Herrera, L. J.; Pomares, H.; Rojas, I.
Resumo: The new technological achievements that have been developed in the last decade are making a change in the profile and behaviour of Internet users and electronic banking. The implementation of this service and the significant increase of its use make mandatory its use for the finance sector and to make investments in order to satisfy and keep the clients by improving their confidence and loyalty. Electronic Bank has become one of the elements that distinguish one bank from another that can make a difference in a certain type of client. The purpose of this paper is to contribute to the research in this important area by applying novel machine learning techniques and algorithms in order to identify client’s profiles as well as asses to apply strategies based on the data that a bank has available. Concretely, starting from a data base, Mutual Information and Delta Test will be applied to identify the most significant variables that determine the global satisfaction with the bank as well as the use of neural networks to predict the satisfaction level.
Palavras-chave: Variable Selection; Mutual Information; Delta Test; RBFNN; e-Bank
Código DOI: 10.21528/CBIC2011-19.1
Artigo em pdf: st_19.1.pdf
Arquivo BibTex: st_19.1.bib