BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Classification of Twitter Accounts into Automated Agents and Human
  Users - Zafar Gilani (Computer Lab)
DTSTART:20170713T140000Z
DTEND:20170713T150000Z
UID:TALK72794@talks.cam.ac.uk
CONTACT:Liang Wang
DESCRIPTION:Online social networks (OSNs) have seen a remark- able rise in
  the presence of surreptitious automated accounts. Massive human user-base
  and business-supportive operating model of social networks\, such as Twit
 ter\, facilitates the creation of automated agents. In this paper we outli
 ne a systematic methodology and train a classifier to categorise Twitter a
 ccounts into ‘automated’ and ‘human’ users. To improve classificat
 ion accuracy we employ a set of novel steps. First\, we divide the dataset
  into four popularity bands to compensate for differences in types of acco
 unts. Second\, we create a large ground truth dataset using human annotati
 ons and extract relevant features from raw tweets. To judge accuracy of th
 e procedure we calculate agreement among human annotators as well as with 
 a bot detection tool. We then apply a Random Forests classifier that achie
 ves an accuracy close to or surpassing human agreement. Finally\, as a con
 cluding step we perform tests to measure the efficacy of our results.
LOCATION:FW26\, Computer Laboratory\, William Gates Building
END:VEVENT
END:VCALENDAR