Research Finds Companies Can Tell Surfers Apart

After the uproar over AOL releasing three months worth of search data, a research paper indicates that online content providers and e-tailers can indeed identify surfers from their browsing history. However, depending on the use of the information, identification could actually help users.

by Kristina Knight

The paper, co-authored by Wharton Professor Balaji Padmanabhan and UC-Davis Professor Catherine Yang, found that every user has a unique “clickprint” that identifies surfing behavior such as page views per session, time spent per page and when pages are visited. After several visits, usually between three and sixteen, this clickprint can differentiate between two users with 100% accuracy. These prints don’t identify users by name, only by habits.

The Wharton research indicated that this identification could be used to prevent Internet fraud, and possibly tailor site sessions to user preferences. Unlike AOL’s release of records, which could have potentially made searchers identifiable by name, a clickprint is research for basic browsing, identifying which pages are viewed the most, when they are viewed and how many viewings are then converted.

Just what this means for Internet privacy is still up in the air, however, the research from Wharton also indicates that clickprints could be used in the future to protect users against identity theft. A clickprint could recognize a user using a stolen credit card based on differences in browsing behavior between the actual card owner and the thief.

The researchers admit more investigation is needed because the ability to create a clickprint does raise privacy concerns, posing a danger to surfers and companies. Meanwhile, one web browser is boasting its ability to keep surfing completely anonymous.