Traditionally, news media organizations used to publish only a few editions of the printed newspapers, and all subscribers of a particular edition used to receive the same information broadcasted by the media organization. The advent of personalized news recommendations has completely changed this simpler news landscape. Such recommendations effectively produce numerous personalized editions of a single newspaper, consisting of only the stories recommended to a particular reader. Although prior works have considered news coverage of different newspapers, due to the difficulty of knowing what news is recommended to whom, there has been no prior study to look into the coverage of information in different personalized news editions. Moreover, the evolution of the effects of personalization on recommended news stories is also not explored. In this work, we make the first attempt to investigate these issues. By collecting extensive data from New York Times personalized recommendations, we compare the information coverage in different personalized editions and investigate how they evolve over time. We observe that the coverage of news stories recommended to different readers are considerably different, and these differences further change with time. We believe that our work will be an important addition to the growing literature on algorithmic auditing and transparency.
[1]
Neil Thurman,et al.
THE FUTURE OF PERSONALIZATION AT NEWS WEBSITES
,
2012
.
[2]
S. Kullback,et al.
Information Theory and Statistics
,
1959
.
[3]
Kevin A. Hassett,et al.
Is newspaper coverage of economic events politically biased?
,
2014
.
[4]
Krishna P. Gummadi,et al.
Dissemination Biases of Social Media Channels: On the Topical Coverage of Socially Shared News
,
2016,
ICWSM.
[5]
Krishna P. Gummadi,et al.
Can Trending News Stories Create Coverage Bias? On the Impact of High Content Churn in Online News Media
,
2015
.
[6]
Loren G. Terveen,et al.
Exploring the filter bubble: the effect of using recommender systems on content diversity
,
2014,
WWW.
[7]
Jianhua Lin,et al.
Divergence measures based on the Shannon entropy
,
1991,
IEEE Trans. Inf. Theory.
[8]
Eli Pariser,et al.
The Filter Bubble: What the Internet Is Hiding from You
,
2011
.
[9]
David Lazer,et al.
Measuring Price Discrimination and Steering on E-commerce Web Sites
,
2014,
Internet Measurement Conference.