Recomendador de billetes en el sector ferroviario

Read full text Summarise with VIRAT
Authors: Hugo Sanz Cuenca
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

En el presente trabajo, se desarrolla el diseño e implementación de un sistema de recomendación de billetes ferroviarios a partir de datos reales de compra, con el objetivo de predecir de forma personalizada el próximo billete de cada usuario, en términos de origen, destino y tipo de producto. A partir de estos datos se realiza un proceso de preprocesado y análisis exploratorio. Posteriormente, se proponen y evalúan distintos modelos, comenzando por una estrategia básica de recomendación y comparando sus resultados globalmente para extraer conclusiones, limitaciones y plantear posibles líneas de mejora. En conjunto, el sistema logra incluir la compra correcta entre las dos primeras recomendaciones para aproximadamente 3 de cada 4 usuarios. ABSTRACT In the present work, the design and implementation of a railway ticket recommender system based on real purchase data are presented, with the aim of predicting each user’s next ticket in a personalised way in terms of origin, destination, and product type. A preprocessing stage and an exploratory analysis are carried out on the data. Subsequently, several models are proposed and evaluated, starting from an initial baseline, and their results are compared to draw conclusions, discuss limitations, and outline possible directions for improvement. Overall, the system includes the actual purchase among the top recommendations for approximately 3 out of every 4 users.

Fields of Study & Subjects

Recommender Systems and Techniques Information Retrieval and Search Behavior Information Retrieval and Data Mining

Publication details

Journal / source
Hispana
Publication date
1 February 2026
Language
ES
Document type
Thesis