dc.contributor |
Universitat Politècnica de Catalunya. Departament d'Arquitectura de Computadors |
dc.contributor |
Meseguer Pallarès, Roc |
dc.contributor.author |
De Oliveira Paegle, Anna Carolina |
dc.date |
2012-12-21 |
dc.identifier.uri |
http://hdl.handle.net/2099.1/17455 |
dc.language.iso |
eng |
dc.publisher |
Universitat Politècnica de Catalunya |
dc.rights |
Attribution-NonCommercial-ShareAlike 3.0 Spain |
dc.rights |
info:eu-repo/semantics/openAccess |
dc.rights |
http://creativecommons.org/licenses/by-nc-sa/3.0/es/ |
dc.subject |
Àrees temàtiques de la UPC::Economia i organització d'empreses::Comerç electrònic |
dc.subject |
Benchmarking (Management) |
dc.subject |
Electronic commerce |
dc.subject |
E-commerce |
dc.subject |
Recommender system |
dc.subject |
Referenciació (Economia) |
dc.subject |
Comerç electrònic |
dc.title |
Benchmarking on web technologies and a recommender system development for E-commerce |
dc.type |
info:eu-repo/semantics/masterThesis |
dc.description.abstract |
Internet has opened a window enabling retailers to sell to anyone, anywhere and at any time. E-commerce has completely changed the way of doing business and in this context appeared the “daily deals” or “group buying” web pages: a business model which attracts millions of customers through the online sale of experiences – like a dinner or a trip – or products with a high percentage of discount, possible due to the great number of buyers.
Motivated by this context of growth, in this project a benchmarking on web functionalities is done for the main group buying web pages and some general electronic marketplaces.
As a result of this benchmarking a web functionality is chosen to be analyzed and developed for an e-commerce deals web: recommender systems. This leads to a second part of this project, where the main techniques to implement a recommendation system are studied and then used to generate a proof of concept for a recommender engine to work in a group buying web environment.
The goal of using a recommendation system in this environment is to make the right information arrive to the right costumers. Recommender systems are valuable both for users and businesses. From a consumer perspective, they may help the users to manage the information overload of the e-commerce world. From a corporate point of view, they may contribute to the cross-sell and upsell of products.
In this project, an item-based collaborative filtering approach is used to generate the recommender engine for the group buying web environment. After the model is designed, a proof of concept focused in the recommender core is implemented. At the end, some evaluation techniques for the recommender system are described. |