HackerPapers

arXiv:cs/0610105cs.CR

How To Break Anonymity of the Netflix Prize Dataset

Arvind Narayanan, Vitaly Shmatikov

Abstract

We present a new class of statistical de-anonymization attacks against high-dimensional micro-data, such as individual preferences, recommendations, transaction records and so on. Our techniques are robust to perturbation in the data and tolerate some mistakes in the adversary's background knowledge. We apply our de-anonymization methodology to the Netflix Prize dataset, which contains anonymous movie ratings of 500,000 subscribers of Netflix, the world's largest online movie rental service. We demonstrate that an adversary who knows only a little bit about an individual subscriber can easily identify this subscriber's record in the dataset. Using the Internet Movie Database as the source of background knowledge, we successfully identified the Netflix records of known users, uncovering their apparent political preferences and other potentially sensitive information.

arXiv comments
Journal reference
DOI
Subjects
cs.CRCryptography and Security cs.DBDatabases

On Hacker News

Linked by 8 identifiable commenters in 8 comments across 8 threads, from Oct 2008 to Apr 2026. Found in 1 HN submission from Nov 2007 to Nov 2007. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments8 threads

  1. UK Biobank health data keeps ending up on GitHub 1 comment

    • 'Anonymisation' schemes are a little like encryption, in that they just get monotonically weaker over time as people work out attacks. But the attacks tend to be much worse. I work in academic open data publishing, and the netflix prize (https://arxiv.org/abs/cs/0610105) hangs…

      shellacView on HNParent context

  2. Ban the sale of precise geolocation 1 comment

    • We should have learned this lesson 20 years ago when researchers were able to deanonymize a lot of the Netflix Prize dataset, which contained nothing except movie ratings and their associated dates. https://arxiv.org/abs/cs/0610105 If movie ratings are vulnerable to…

      teraflopView on HNParent context

  3. Yt-dlp: Upcoming new requirements for YouTube downloads 1 comment

    • > I'm telling you that Firefox is going to be out of business soon Do you not think everyone saying Firefox is going to be out of business soon plays a role in this? Regardless, I think you've ignored the root of my argument. I'm not trying to be a Firefox fanboy here but it's…

      godelskiView on HNParent context

  4. California moves to silence Stanford researchers who got data to study education 1 comment

    • LOL. I'm not going to deanonymize datasets for an internet stranger. Certainly not for free, but also not at any hourly rate unless I know that the organization employing me either has the original non-anonymized dataset or at least have very strict internal controls about how…

      ke88yView on HNParent context

  5. Netflix tweet raises privacy questions 1 comment
  6. Ask HN: Papers you read in 2015? 1 comment

    • It didn't get published this year but I though Robust De-Anonymization of Large Datasets - http://arxiv.org/pdf/cs/0610105.pdf

      roninbView on HNParent context

  7. $3 million machine learning prize 1 comment

    • You're perhaps right. Of course, these people are on things like Facebook, Twitter and blogs and if you can narrow down age, gender, location and medical condition, you might be able to correlate with public posts. You can also look for specific people in there if you have…

      indigovioletView on HNParent context

  8. "Anonymous" Facebook Dataset Is Identified as Harvard Students 1 comment

    • It is getting easy to break anonymous datasets with mining: Netflix: http://arxiv.org/abs/cs/0610105 AOL: http://www.techcrunch.com/2006/08/07/aol-this-was-a-screw-up...

      gaikaView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.