Emma Lurie twitter profile CV link


Emma Lurie I'm a second year PhD student at the UC Berkeley School of Information where I am advised by Deirdre Mulligan. My current research interests include online reputation management and reputation repair, automated and crowdsourced fact-checking, and election-related search engine audits.

I have a BA in Computer Science and Chinese Language & Culture from Wellesley College. Previously, I worked as an undergraduate research assistant at the Wellesley College Cred Lab and the MIT Election Data and Science Lab.

prounouns: she/her



Peer Reviewed Papers


  • Search engine audits of election related information on Google primarily query candidate names (e.g. Elizabeth Warren, Donald Trump).
  • We find that voters search for many queries not typically tracked in audit studies.
  • Political science has a rich literature that can be used to design future audit studies that better model voter search behavior.

Opening Up the Black Box: Auditing Google’s Top Stories Algorithm. Emma Lurie and Eni Mustafaraj. AAAI FLAIRS 2019.

  • The Google Top stories feature news coverage as presented in Top stories relies on a small number of mainstream publishers.
  • Google Top stories appears to be experimenting with lesser known sources in the third position of the panel.
  • Even through Google's algorithmic lens, we can see media outlets' "selection bias" (e.g. certain publishers are more likely to cover particular topics).
  • Web literacy best practices urge individuals to examine third-party information to evaluate the credibility of source.
  • We find a low prevelance of high quality Google search results pages for local newspapers. This is problematic in part because a common disinformation tactic is for a site to appear to be a local newspaper.
  • We observed 30 students as they interacted with the Google SERP to identify which parts were helpful in evaluating the credibility of sources. Overall, the Google curated features (Knowledge Panel, Top stories) were cited as particularly useful by participatns.

Non-Archival Publications


Investigating Causal Effects of Instructions in Crowdsourced Claim Matching. Computation + Journalism 2020. With Lucy Li, Sofia Dewar, Masha Belyi, Daniel Rincón, John Baldwin, and Rajvardhan Oak.
Considering Contestability in Automated Fact-Checking Systems. Contestability Workshop at CSCW 2019.
The Challenges of Algorithmically Assigning Fact-checks: A Sociotechnical Examination of Google's Reviewed Claims. Undergraduate Thesis, 2019.
How the Interplay of Google and Wikipedia Affects Perceptions of Online News Sources. With Annabel Rothschild and Eni Mustafaraj. Computation + Journalism 2019.

Blogging

What is “good enough” for automated fact-checking?
Googling for the Kansas Primary
Why Google Isn't Always Right