Propuesta de implementación de técnicas de clusterización para la recomendación personalizada de productos para una plataforma de e-commerce en Honduras
Cargando...
Fecha
2025-06-20
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Tecnológica Centroamericana UNITEC
Resumen
Esta tesis tiene como propósito mejorar la personalización de las recomendaciones de productos en la plataforma LynxPay en Honduras, donde el comercio electrónico está en crecimiento, pero enfrenta desafíos para personalizar la experiencia de compra. El objetivo es segmentar usuarios en grupos homogéneos mediante algoritmos de clusterización como K-Means y DBSCAN permitiendo una mejor implementación de técnicas de recomendación, basándose en su comportamiento. Para ello, se recopilaron y analizaron datos de la plataforma, se aplicaron los algoritmos y se evalúo su efectividad mediante métricas como la tasa de conversión y la retención de clientes. Se analizó la viabilidad económica considerando el costo-beneficio. Finalmente, se espera que los resultados de este estudio contribuyan a optimizar las estrategias de marketing y ventas en el comercio electrónico hondureño.
This thesis aims to improve the personalization of product recommendations on the LynxPay platform in Honduras, where e-commerce is growing, but faces challenges to personalize the shopping experience. The objective is to segment users into homogeneous groups using clustering algorithms such as K-Means and DBSCAN, allowing a better implementation of recommendation techniques, based on their behavior. To do this, data from the platform was collected and analyzed, algorithms were applied, and its effectiveness was evaluated using metrics such as conversion rate and customer retention. Economic viability was analyzed considering cost-benefit. Finally, it is expected that the results of this study will contribute to optimizing marketing and sales strategies in Honduran e-commerce.
This thesis aims to improve the personalization of product recommendations on the LynxPay platform in Honduras, where e-commerce is growing, but faces challenges to personalize the shopping experience. The objective is to segment users into homogeneous groups using clustering algorithms such as K-Means and DBSCAN, allowing a better implementation of recommendation techniques, based on their behavior. To do this, data from the platform was collected and analyzed, algorithms were applied, and its effectiveness was evaluated using metrics such as conversion rate and customer retention. Economic viability was analyzed considering cost-benefit. Finally, it is expected that the results of this study will contribute to optimizing marketing and sales strategies in Honduran e-commerce.
Palabras clave
Clusterización, Comercio electrónico, Recomendaciones personalizadas, LynxPay