Recommender Systems in E-News Environments: Challenges and Problems Faced

Postgraduate Thesis uoadl:2897521 200 Read counter

Unit:
Κατεύθυνση Ψηφιακά Μέσα Επικοινωνίας και Περιβάλλοντα Αλληλεπίδρασης
Library of the Faculties of Political Science and Public Administration, Communication and Mass Media Studies, Turkish and Modern Asian Studies, Sociology
Deposit date:
2020-02-22
Year:
2020
Author:
Kalathas Iasonas
Supervisors info:
Μουρλάς Κωνσταντίνος, Αναπληρωτής Καθηγητής, Τμήμα Επικοινωνίας και Μέσων Μαζικής Ενημέρωσης, Εθνικό και Καποδιστριακό Πανεπιστήμιο της Αθήνας
Original Title:
Συστήματα παραγωγής συστάσεων σε περιβάλλοντα ηλεκτρονικών ειδήσεων: Προκλήσεις και Προβλήματα που συναντάμε
Languages:
Greek
Translated title:
Recommender Systems in E-News Environments: Challenges and Problems Faced
Summary:
The scope of this thesis was to explore the recommender systems that are utilized in E-News Environments. Our goal was to highlight the most important challenges faced in that field and to present a proposal of a more user-centered design of a recoomender systems of e-news.
At first, in order to comprehend recommender systems, we analysed the most prominent parameters and the essential stages of the way recommender systems operate.
Moreover, we investigated the related literature to understand and negotiate challenges and problems faces such as the cold-start problem, data sparsity as well as quality factors like diversity, novelty and serendipity. The findings contributed to the formation of a design proposal of a news recommender system that takes into account the users' needs and explicit contributions emphasizing in factors like diversity of the news content, discovery of unexplored news categories, pluralism of standpoints, escapism and deepening of knowledge for certain news categories.
Main subject category:
Social, Political and Economic sciences
Other subject categories:
Technology - Computer science
Keywords:
Recommender Systems, E-News Environments, Algorithmic Transparency, Diversity, Serendipity, User-Based Content Personalisation
Index:
No
Number of index pages:
0
Contains images:
Yes
Number of references:
92
Number of pages:
103
File:
File access is restricted only to the intranet of UoA.

Διπλωματική Εργασία Ι. Καλαθάς.pdf
1 MB
File access is restricted only to the intranet of UoA.