Are you my friend? [Twitter response estimator]

• Published in: 2010 7th IEEE Consumer Communications and Networking Conference pp. 1 - 5
• Main Authors: Baatarjav, E.-A., Amin, A., Dantu, R., Gupta, N.K.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-01-2010
• Subjects: Computer science; Facebook; MySpace; Network servers; Privacy; Social network services; Twitter
• ISBN: 9781424451753; 1424451752
• ISSN: 2331-9852
• DOI: 10.1109/CCNC.2010.5421732

With Twitters growing popularity, privacy has become a major concern for users wary of sensitive information falling in the wrong hands. A typical Twitter user carries hundreds of followers people who have subscribed to users twitter feeds. Our goal is to target followers that a Twitter user deems safe. Therefore, selecting followers with a closer relationship could help decrease the risk of sensitive information being sent to unknown people. We propose a privacy management system that helps a Twitter user restrict information to only certain followers based on the strength of their social tie. This system would incorporate two tools: the Exclusivity meter and the Twitter Response Estimator. The Exclusivity Meter employs user's activity profile to guess similarities between users. Preliminary results have indicated that similarity in time and level of activity between user and follower does suggest a stronger social tie. The Twitter Response Estimator uses a measure of prestige to gauge the probability of response. When applied to a set of followers, the estimator separates real friends from recreational followers.