Please use this identifier to cite or link to this item:
Title: Desenvolupament d'un model predictiu del preu dels allotjaments turístics de Barcelona
Author: Manresa Rigo, Pere Antoni
Director/Tutor: Vitrià i Marca, Jordi
Keywords: Mètodes estadístics
Dades massives
Treballs de fi de grau
Teoria de la predicció
Statistical methods
Big data
Computer software
Prediction theory
Bachelor's thesis
Issue Date: 6-Jun-2016
Abstract: This project carries out a development of a statistical modelling research which intends to represent the most important features of Airbnb users’ usage of its services. This research comprises of a deployment of different statistical learning techniques, from building lineal regression models, and implementing ensemble algorithms such as Gradient Boosting Regressor, to apply state-of-the-art optimization methods. The goal of this project is not to create a deep and flexible statistical model by deploying complex algorithms but leverage a wide range of simpler techniques in order to build a stronger and more comprehensive model. The innovative market of software products within the touristic sector has been one of the main targets of many actual companies. The Airbnb methodology’s real battle against many old-fashioned accommodation services has raised the interest of many housing companies which are investing a lot of money in understanding their secrets. The illustration of this project’s results will help readers to achieve a better understanding of how Airbnb is actually used by its users, and how they could obtain a compensating revenue out from its services.
Note: Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2016, Director: Jordi Vitrià i Marca
Appears in Collections:Programari - Treballs de l'alumnat
Treballs Finals de Grau (TFG) - Enginyeria Informàtica

Files in This Item:
File Description SizeFormat 
memoria.pdfMemòria7.51 MBAdobe PDFView/Open
codi_font.zipCodi font26.52 MBzipView/Open

This item is licensed under a Creative Commons License Creative Commons