Sistema de recomendación integrado en ChatGPT para analizar el comportamiento poscompra de usuarios de una tienda e-commerce

• Autor: Ovalle, Christian
• Formato: objeto de conferencia
• Fecha de Publicación: 2024
• Institución: Universidad Tecnológica del Perú
• Repositorio: UTP-Institucional
• Lenguaje: español
• OAI Identifier: oai:repositorio.utp.edu.pe:20.500.12867/14350
• Enlace del recurso: https://hdl.handle.net/20.500.12867/14350; https://doi.org/10.18687/LACCEI2024.1.1.227
• Nivel de acceso: acceso abierto
• Materia: Recommendation System; ChatGPT; E-commerce; Post-purchase; https://purl.org/pe-repo/ocde/ford#2.02.04

Recommender systems have had a great development in recent years, helping exponentially in the electronic commerce sector. This has many applications to improve user behavior factors with different filtering techniques; however, most of these systems lack a presentation and interaction model that really influences users. In this context, e-commerce sites seek different strategies to allocate recommendations viewed by the online user in an accurate and timely manner; even so, reviewing different articles it is not very clear if the way in which recommended articles are presented has a positive impact on user behavior. On the other hand, the technology of conversational artificial intelligence systems had a great size, highlighting ChatGPT as an innovative tool. Finally, this research seeks to validate whether the implementation of an integrated SR in ChatGPT influences the post-purchase behavior of users of an ecommerce store. The results show that by taking advantage of the potential of conversational AI to provide more effective and personalized recommendations, there is an increase of 34.15% with respect to the recommendation of users, while in the purchase of recommended products there is an exponential increase of 54.05%; Likewise, it is evident that users who make repurchases after 14 days from their initial purchase have an increase of 46.67%; finally, that the repurchase of products from the ecommerce store has a slight significant increase of 9.52%.