BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Mining Large-Scale Internet Data to Find Stealthy Abuse - Mobin Ja
 ved\, ICSI Berkeley
DTSTART:20171107T130000Z
DTEND:20171107T140000Z
UID:TALK93997@talks.cam.ac.uk
CONTACT:Gemma Gordon
DESCRIPTION:Internet abuse is advancing to a hard-to-detect stealthy space
 . A number of factors contribute to this shift: increasing sophistication 
 of adversaries in response to maturing Internet defenses\, new powerful ad
 versaries (such as nation-state actors) surfacing\, and the emergence of a
 n underground economy that facilitates access to the tools and resources r
 equired to conduct attacks. Further\, the shift towards high-speed network
 s plays to the advantage of abusers\, producing data of a nature and scale
  that serves as another obfuscation layer for their abuse operations. From
  defenders’ point of view\, the detection task is hard: the threat signa
 l is often buried inside a sea of benign data. In this talk\, I will discu
 ss my work on deriving actionable security intelligence from hundreds of m
 illions of log records. I will begin with an overview on Internet abuse re
 search and will discuss in depth my work on two detection problems: (i) de
 tecting large-scale coordinated and stealthy attacks\, and (ii) mining net
 work traffic to find surreptitious forms of online tracking.\n\nBio: Mobin
  Javed is a Post-doctoral Research Scholar in the Networking and Security 
 group at the International Computer Science Institute\, Berkeley. She rece
 ived her Ph.D. from UC Berkeley in 2016 advised by Vern Paxson\, and will 
 be joining LUMS as an Assistant Professor in Spring 2018. Her research foc
 uses on analyzing real-world data from large-scale networked systems to un
 derstand Internet adversaries\, and to develop practically deployable solu
 tions for fighting cyber threats. Some of her projects include: (i) detect
 ion of stealthy and coordinated attacks\, (ii) measurement of surreptitiou
 s tracking\, and (iii) measurement and evasion of Internet censorship. Her
  work on detecting credential spear-phishing attacks is the winner of the 
 2017 Internet Defense Prize. She also has a keen interest in social impact
 \, and was recently selected as a fellow at the Data Science for Social Go
 od (DSSG) program at the University of Chicago\, where she worked with the
  government of Mexico to help fight poverty through data science.  Mobin i
 s also the co-founder of GradApp Lab\, Pakistan\, a mentoring effort that 
 connects aspiring grad school applicants with mentors abroad.
LOCATION:Computer Laboratory\, William Gates Building\, Room FW11
END:VEVENT
END:VCALENDAR