Please use this identifier to cite or link to this item:

http://hdl.handle.net/10609/63685
Title: Desarrollo de un recomendador de productos basado en Extreme Gradient Boosting
Author: López Serrano, Pablo
Director: Kanaan Izquierdo, Samir
Tutor: Ventura Royo, Carles  
Others: Universitat Oberta de Catalunya
Keywords: Gradient boosting
prediction
recommender
Issue Date: 31-May-2017
Publisher: Universitat Oberta de Catalunya
Abstract: In this final project has been developed a product classifier algorithm using Extreme Gradient Boosting technique. Santander Product Recommendation is the problem to be solved selected from Kaggle platform. In this problem the competitor has to predict the seven products with higher probability to be purchased in the future. The purpose of this project is to resolve a real predictive problem with higher data and apply one of the most success predictive method in the competitions of Machine Learning. Further, at the end of the resolution, the author will compare the results with the best users of the competition.
Language: Spanish
URI: http://hdl.handle.net/10609/63685
Appears in Collections:Bachelor thesis, research projects, etc.

Share:
Export:
Files in This Item:
File Description SizeFormat 
Enlaces de archivos csv.txt335 BTextView/Open
CodigoSubmissionsMix.R2.26 kBUnknownView/Open
CodigoFinalSubmission.R2.52 kBUnknownView/Open
Codigo.R9.72 kBUnknownView/Open
plopezseTFM0617memoria.pdfMemoria del trabajo fin de máster1.96 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons