Beyond Filter Bubbles: News Recommendation on Google Search
by Eni Mustafaraj, Computer Science, Wellesley College
A few years ago, the term “filter bubble” entered our discourse through Eli Pariser’s TED talk, highlighting problematic Google search results. Are filter bubbles still present? How do algorithmic news curation systems, like Google’s Top stories, choose what articles to recommend daily to their audiences, by promoting some articles over others? Google has stated that its goal is to connect people with high quality news from a variety of perspectives, but it is also vague about how it resolves the tension between high quality news and politically (hyper)partisan perspectives. In this talk, I will discuss methods for auditing search engines, what I have learned from auditing Google’s Top stories results about political events in the past two years, and directions for future interdisciplinary research.
about the speaker
Dr. Eni Mustafaraj is an Assistant Professor of Computer Science at Wellesley College. Her research focuses on methods for exposing misinformation and manipulation in sociotechnical platforms. She has been auditing Google search results in the context of US elections since 2008, and discovered the first use of political Twitter bots for spreading misinformation during elections in 2010.