PICAE – Intelligent publication of audiovisual and editorial contents
Visualitza/Obre
Estadístiques de LA Referencia / Recolecta
Inclou dades d'ús des de 2022
Cita com:
hdl:2117/356972
Tipus de documentProjecte Final de Màster Oficial
Data2021-06-30
Condicions d'accésAccés obert
Tots els drets reservats. Aquesta obra està protegida pels drets de propietat intel·lectual i
industrial corresponents. Sense perjudici de les exempcions legals existents, queda prohibida la seva
reproducció, distribució, comunicació pública o transformació sense l'autorització del titular dels drets
Abstract
The development in internet infrastructure and technology in last tow decades have given users and retailers the possibility to purchase and sell items online. This has of course broadened the horizons of what products can be offered outside of the traditional trading sense, to the point where virtually any product can be offered. These massive online markets have had a considerable impact on the habits of consumers, providing them access to a greater variety of products and information on these goods. This variety has made online commerce into a multi-billion dollar industry but it has also put the customer in a position where it is getting increasingly difficult to select the products that best fit their individual needs. In the same vein, the rise of both availability and the amounts of data that computers have been able to process in the last decades have allowed for many solutions that are computationally expensive to exist, and recommender systems are no exception. These systems are the perfect tools to overcome the information overload problem since they provide automated and personalized suggestions to consumers. The PICAE project tackles the recommendation problem in the audiovisual sector. The vast amount of audiovisual content that is available nowadays to the user can be overwhelming, which is why recommenders have been increasingly growing in popularity in this sector ---Netflix being the biggest example. PICAE seeks to provide insightful and personalized recommendations to users in a public TV setting. The PICAE project develops new models and analytical tools for recommending audiovisual and editorial content with the aim of improving the user experience, based on their profile and environment, and the level of satisfaction and loyalty. These new tools represent a qualitative improvement in the state of the art of television and editorial content recommendation. On the other hand, the project also improves the digital consumption index of these contents based on the identification of products that these new forms of consumption demand and how they must be produced, distributed and promoted to respond to the needs of this emerging market. The main challenge of the PICAE project is to resolve two differentiating aspects with respect to other existing solutions such as: variety and dynamic contents that requires a real-time analysis of the recommendation and the lack of available information about the user, who in these areas is reluctant to register, making it difficult to identify in multi-device consumption. This document will explain the contributions made in the development of the project, which can be divided in two: the development of the project, which can be divided in two: the development of a recommender system that takes into account information of both users and items and a deep analysis of the current metrics used to assess the performance of a recommender system.
MatèriesRecommender systems (Information filtering), Factorization (Mathematics), Sistemes recomanadors (Filtratge d'informació), Factorització (Matemàtica)
TitulacióMÀSTER UNIVERSITARI EN INNOVACIÓ I RECERCA EN INFORMÀTICA (Pla 2012)
Col·leccions
Fitxers | Descripció | Mida | Format | Visualitza |
---|---|---|---|---|
158585.pdf | 4,167Mb | Visualitza/Obre |