Authors:
Mahsa Badami
1
;
Olfa Nasraoui
1
and
Patrick Shafto
2
Affiliations:
1
Knowledge Discovery and Web Mining Lab, Computer Science and Computer Engineering Department, University of Louisville, Louisville, KY and U.S.A.
;
2
Department of Mathematics and Computer Science, Rutgers University - Newark, Newark, NJ and U.S.A.
Keyword(s):
Recommender System, Polarization, Controversy, Big data, Algorithmic Bias.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Collaborative Filtering
;
Computational Intelligence
;
Evolutionary Computing
;
Interactive and Online Data Mining
;
Knowledge Discovery and Information Retrieval
;
Knowledge-Based Systems
;
Machine Learning
;
Pre-Processing and Post-Processing for Data Mining
;
Soft Computing
;
Symbolic Systems
;
User Profiling and Recommender Systems
Abstract:
Personalized recommender systems are commonly used to filter information in social media, and recommendations are derived by training machine learning algorithms on these data. It is thus important to understand how machine learning algorithms, especially recommender systems, behave in polarized environments. We investigate how filtering and discovering information are affected by using recommender systems. We study the phenomenon of polarization within the context of the users interactions with a space of items and how this affects recommender systems. We then investigate the behavior of machine learning algorithms in such environments. Finally we propose a new recommendation model based on Matrix Factorization for existing collaborative filtering recommender systems in order to combating over-specialization in polarized environments toward counteracting polarization in human-generated data and machine learning algorithms.