Predicting the success of banking telemarketing through the use of decision trees

• Autores: Ventura Ramos, Rony Tito, Jacobo Castillo, Andrew Pold, Begazo Ticona, Jesus, Gomez Velasco, Brian Jhosep
• Formato: artículo
• Fecha de Publicación: 2023
• Institución: Universidad La Salle
• Repositorio: Revistas - Universidad La Salle
• Lenguaje: español
• OAI Identifier: oai:ojs.revistas.ulasalle.edu.pe:article/84
• Enlace del recurso: https://revistas.ulasalle.edu.pe/innosoft/article/view/84; https://doi.org/10.48168/innosoft.s11.a84; https://purl.org/42411/s11/a84; https://n2t.net/ark:/42411/s11/a84
• Nivel de acceso: acceso abierto
• Materia: Telemarketing; Decision trees; Artificial Intelligence; Árboles de decisión; Inteligencia artificial

Telemarketing is an interactive direct marketing technique in which a telemarketing agent solicits potential customers over the phone to make a sale of merchandise or a service. One of the great problems of telemarketing is to specify the list of clients that presents a greater probability of buying the product that is offered. In this article, we propose a personalized decision support system that can automatically predict the decision of the target audience after making a telemarketing call, in order to increase the effectiveness of direct advertising campaigns and consequently reduce the cost and cost. campaign time. The artificial intelligence method used in this work is the decision tree evaluated with the metrics of precision, accuracy and completeness. After applying the artificial intelligence method we obtain an accuracy, precision and completeness greater than 80%. The conclusions reached by the team are that in order to improve the decision tree model it is important to carry out a prior analysis of the data using statistical techniques or diagrams, to obtain a reference to the data and apply balancing techniques to obtain the best possible model.