1 / 36

Take This Personally: Pollution Attacks on Personalized Services

22 nd USENIX Security (August, 2013). Xinyu Xing, Wei Meng, Dan Doozan , Georgia Institute of Technology Alex C. Snoeren , UC San Diego Nick Feamster , and Wenke Lee, Georgia Institute of Technology. Take This Personally: Pollution Attacks on Personalized Services. Outline.

khan
Télécharger la présentation

Take This Personally: Pollution Attacks on Personalized Services

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. 22nd USENIX Security (August, 2013) Xinyu Xing, Wei Meng, Dan Doozan, Georgia Institute of Technology Alex C. Snoeren, UC San Diego Nick Feamster, and WenkeLee, Georgia Institute of Technology Take This Personally: Pollution Attacks on Personalized Services

  2. Outline • Introduction • Overview and Attack Model • Pollution Attacks on YouTube • Google Personalized Search • Pollution Attacks on Amazon A Seminar at Advanced Defense Lab

  3. Introduction • Modern Web services are increasingly relying upon personalizationto improve the quality of their customers’ experience. • Many services with personalized content log their users’ Web activities. A Seminar at Advanced Defense Lab

  4. This paper... • We demonstrate that contemporary personalization mechanisms are vulnerable to exploit. A Seminar at Advanced Defense Lab

  5. Our Attack • We show that YouTube, Amazon, and Google are all vulnerable to the same class of cross-site scripting attack, which we call a pollution attack, that allows third parties to alter the customized content. • A distinguishing feature of our attack is that it does not exploit any vulnerability in the user’s Web browser. A Seminar at Advanced Defense Lab

  6. Overview and Attack Model • The main instrument that a service provider can use to affect the content that a user sees is modifying the choice set. • When a user issues a query, a service’s personalization algorithmaffects the user’s choice set for that query. A Seminar at Advanced Defense Lab

  7. Overview and Attack Model (cont.) • In this paper, we focus on how changes to a user’s history can affect the choice set, holding other factors fixed. • This attack requires three steps: • Model the service’s personalization algorithm. • Create a “seed” to pollute the user’s history. • Inject the seed with a vector of false clicks. A Seminar at Advanced Defense Lab

  8. A Seminar at Advanced Defense Lab

  9. Pollution Attacks on YouTube • Personalization rule • Consider only those videos that the user watched for a long period of time • Similar viewing histories • Notrecommend a video the user has already watched • Two of suggested videos are recommended based upon personalization A Seminar at Advanced Defense Lab

  10. A Seminar at Advanced Defense Lab

  11. Preparing Seed Videos Video channel (C) ΩS ΩT A Seminar at Advanced Defense Lab

  12. Inject Seed Videos • We see the video: • http://www.youtube.com/user_watch?plid=<value>&video_id=<value> • We watch for a period of time: • http://www.youtube.com/set_awesome?plid=<value>&video_id=<value> A Seminar at Advanced Defense Lab

  13. Experimental Design A Seminar at Advanced Defense Lab

  14. Evaluation • We evaluated the effectiveness of our pollution attacks by logging in as the victim user and viewing 114representative videos. A Seminar at Advanced Defense Lab

  15. Evaluation (New Accounts) • Successfully • we computed • the Pearson correlation between the showing frequencies and the lengths of the target videos • 0.54 => medium • the Pearson correlation between the showing frequencies and the view counts of the target videos • 0.23 => moderate A Seminar at Advanced Defense Lab

  16. A Seminar at Advanced Defense Lab

  17. A Seminar at Advanced Defense Lab

  18. Evaluation (Existing Accounts) • For existing channel OnlyyouHappycamp • 14 of the 22 volunteers (64%) • Ten of our volunteers shared their histories • The majority of the videos recommended to users for whom our attacks have low promotion rates have longer lengths and more view counts than our target videos. A Seminar at Advanced Defense Lab

  19. A Seminar at Advanced Defense Lab

  20. Google Personalized Search • We describe two classes of personalization algorithms: • contextual personalization • persistent personalization A Seminar at Advanced Defense Lab

  21. A Seminar at Advanced Defense Lab

  22. Identifying Search Terms • Contextual Personalization • The keywords injected into a user’s search history should be both relevant to the promoting keywordand unique to the website being promoted. A Seminar at Advanced Defense Lab

  23. Identifying Search Terms (cont.) • Persistent Personalization • In this case, the size of the keyword set should be larger than that used for a contextual attack in order to have a greater effect on the user’s search history. • An attacker can safely inject roughly 50 keywords a minute using cross-site request forgery. • we assume an attacker can inject at most 25keywords into a user’s profile A Seminar at Advanced Defense Lab

  24. Contextual Personalization Google results URLs having unique <meta> keywords 30 URLs 5,761 Search Terms from made-in-china.com 30 URLs URLs having unique <meta> keywords 30 URLs 30 URLs URLs having unique <meta> keywords 1,739 search terms 151,363 unique URLs 2,136 URLs A Seminar at Advanced Defense Lab

  25. 2,136 URLs for Contextual Personalization A Seminar at Advanced Defense Lab

  26. Persistent Personalization Google results URLs having unique Google AdWords keywords 30 URLs 551 Search Terms from made-in-china.com 30 URLs 30 URLs 30 URLs 151,363 unique URLs 15,979 URLs A Seminar at Advanced Defense Lab

  27. Evaluation • Contextual Personalization 44% 1.1% 62.8% 28% A Seminar at Advanced Defense Lab

  28. Evaluation (cont.) • Persistent Personalization 17% 4.3% 22.7% ??% A Seminar at Advanced Defense Lab

  29. Evaluation (cont.) • Real Users • 97.1% of our 729 previously successful contextual attacks remain successful. • Only 77.78% of the persistent pollution attacks that work on fresh accounts achieve similar success A Seminar at Advanced Defense Lab

  30. Pollution Attacks on Amazon • Amazon tailors a customer’s homepage based on the previous purchase, browsing and searching behavior of the user. • We focused on the personalized recommendations Amazon generates based on the browsing and searching activities A Seminar at Advanced Defense Lab

  31. A Seminar at Advanced Defense Lab

  32. Amazon Recommendations • Amazon’s personalization is based on history that maintained by the user’s web browser. • Session cookie A Seminar at Advanced Defense Lab

  33. Identifying Seed Products and Terms • Visit-Based Pollution • the attacker visits the Amazon page of the product and retrieves the related products that are shown on Amazon page of the targeted product. • Search-Based Pollution • An attacker could use a natural language toolkit to automatically extract a candidate keyword set from the targeted product’s name. A Seminar at Advanced Defense Lab

  34. A Seminar at Advanced Defense Lab

  35. A Seminar at Advanced Defense Lab

  36. Q & A A Seminar at Advanced Defense Lab

More Related