I should not recommend it to you even if you will like it: The ethics of recommender systems
Document Type
Article
Publication Date
1-2-2016
Abstract
In this paper, we extend the current research in the recommendation system community by showing that users did attach ethical consideration to items. In an experiment (N = 111) that manipulated several moral factors regarding the potentially harmful content in movies, books, and games, users were asked to evaluate the appropriateness of recommending these items to teenagers and adult couples. Results agreed with previous studies in that gender plays a key role in making moral judgment, especially regarding the ethical appropriateness of an item. The pilot study further identifies degrees of aversion regarding the appeal of these elements in media for ethical recommendations. Based on the study, we propose a user-initiated ethical recommender system to help users pick up morally appropriate items during the post-recommendation process. We believe that the ethical appropriateness of items perceived by end users could predict the trust and credibility of the system.
Publication Title
New Review of Hypermedia and Multimedia
First Page Number
111
Last Page Number
138
DOI
10.1080/13614568.2015.1052099
Recommended Citation
Tang, Tiffany Ya and Winoto, Pinata, "I should not recommend it to you even if you will like it: The ethics of recommender systems" (2016). Kean Publications. 1764.
https://digitalcommons.kean.edu/keanpublications/1764