Investigating Facebook groups through a random graph model
Version 2 2024-06-04, 02:37Version 2 2024-06-04, 02:37
Version 1 2014-10-28, 10:23Version 1 2014-10-28, 10:23
journal contribution
posted on 2014-01-01, 00:00authored byDinithi Pallegedara, Lei PanLei Pan
Facebook disseminates messages for billions of users everyday. Though there are log files stored on central servers, law enforcement agencies outside of the U.S. cannot easily acquire server log files from Facebook. This work models Facebook user groups by using a random graph model. Our aim is to facilitate detectives quickly estimating the size of a Facebook group with which a suspect is involved. We estimate this group size according to the number of immediate friends and the number of extended friends which are usually accessible by the public. We plot and examine UML diagrams to describe Facebook functions. Our experimental results show that asymmetric Facebook friendship fulfills the assumption of applying random graph models.
History
Journal
Journal of multimedia
Volume
9
Issue
1
Pagination
25 - 34
Publisher
Academy Publisher
Location
Finland
ISSN
1796-2048
Language
eng
Publication classification
C1 Refereed article in a scholarly journal; C Journal article