Deep Learning Techniques on Recommender Systems

Graduate Thesis uoadl:2961302 169 Read counter

Unit:
Department of Informatics and Telecommunications
Πληροφορική
Deposit date:
2021-09-24
Year:
2021
Author:
STOIKOY KONSTANTINA
Supervisors info:
Παναγιώτης Σταματόπουλος, επίκουρος καθηγητής, Τμήμα Πληροφορικής και Τηλεπικοινωνιών, ΕΚΠΑ
Original Title:
Deep Learning Techniques on Recommender Systems
Languages:
English
Translated title:
Deep Learning Techniques on Recommender Systems
Summary:
A recommender system is a tool that filters information and suggests content to users which is relevant to their interests. Recommender systems have seen a rise in their use in the recent years as a result of the increasing internet use which provides researchers with huge amounts of user data. The purpose of this thesis is to study the various techniques that are applied to recommender systems as well as the deep learning models that are used to enhance those systems. Moreover, the evaluation methods of the recommender systems are described along with the challenges they face. Followingly, an implementa­ tion of a recommender system for video games which employs deep learning algorithms is provided followed by the interpretation of the results. At the end, some concerns and suggestions about the future in the field of recommendations are mentioned.
Main subject category:
Technology - Computer science
Keywords:
recommender system, recommendations, deep learning, machine learn­ing, artificial intelligence, challenges, steam, games
Index:
Yes
Number of index pages:
5
Contains images:
Yes
Number of references:
51
Number of pages:
49
thesis.pdf (4 MB) Open in new window

 


code.zip
113 KB
File access is restricted.