Wil Haywood Rossana Bua Rhishi Katoch Kelvyn AraujoValdez Steve Kim Enormously popular Estimated 23 of all US Adults 71 stay in touch with current friends 50 connect to old friends whove lost touch ID: 745105
Download Presentation The PPT/PDF document "Social Network Forensic Tools" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Social Network Forensic Tools
Wil
Haywood
Rossana Bua
Rhishi Katoch
Kelvyn
Araujo-Valdez
Steve KimSlide2
Enormously popular
Estimated 2/3 of all U.S. Adults*
71% stay in touch with current friends50% connect to old friends who’ve lost touchEstimated 98% of all 18-24 year old who are online and access social media each month**75% of online senior adults engage82% of adults ages 55 to 64*Pew Research April-May 2011**Experian Simmons Report, 2011
Social Networking SitesSlide3
Facebook and Security IssuesFacebook
Most
popular of all social media sites800 million users as of December 2011More than 50% of all users log in dailyAverage user has 130 friendsCybercrimeInternet FraudPhishingSlide4
Forensic ToolsIs there a way for us to remove the anonymity of users in social media sites such as Facebook? Slide5
Facial Recognition in Use
Carnegie
Mellon University conducted experiments to see if an individual could be identified based on the availability of select information, both online and offline.They searched public profile images from online social networks and compared them to dating sites to determine how accurate the software was in identificationThey collected images of consenting students walking on a campus, and compared those to Facebook. Students provided some information and were asked to identify themselves in offline images.Researchers gathered personal information previously unknown from a face by training an algorithm to identify Facebook profile owners.Slide6
Face.comWe use face.com application to identify small sample of users on Facebook. Slide7
Detect an image
How to get started…
We created an account We created a new application Select method face.detect Upload a picture or urlCall methodAnswer Image detects five points on face with different attributes Glasses, Smiling, Face, Gender, Mood, Lips Slide8
Save tagged information
To name a specific person (based on known information):
Change method to save.tagEnter the stored tag idLook for the user of a known account.Slide9
Train the API to recognize a face
Providing a username for a face trains the API to recognize that known user on FacebookSlide10
Multi-identification
Can the API identify a distinct face among many?
Slide11
Multi-identificationSlide12
More than one ID stored in API
Multiple Facebook user IDs are entered into the API.Slide13
Facial Recognition Concerns
Potential for misuse – Privacy issue raised
:Sen. Jay Rockefeller: “While there may be great potential for commercial, personal, and law enforcement users of this emerging technology, its development also raises numerous questions about individual privacy”Slide14
Any Questions?