Browse Prior Art Database

Method and System for Detecting Anomalous Events and Behaviors in Social Networks

IP.com Disclosure Number: IPCOM000235857D
Publication Date: 2014-Mar-27
Document File: 3 page(s) / 141K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system is disclosed for detecting anomalous events and behaviors in social networks.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 3

Method and System for Detecting Anomalous Events and Behaviors in Social

Networks

Disclosed is a method and system for detecting anomalous events and behaviors in social networks. The method and system performs several analytics on textual posts and behavior on the social networks to help users examine a situation and make decision on possible anomalous events and behaviors that are prone to occur . Wide range of analytic results are displayed to the users on a particular interface to monitor and give feedback on the results. Optionally, users can also view details of the events or individuals involved in anomalous behavior around the anomalous events .

In accordance with the method and system, a sequence of steps are performed in order to display a progress of anomalous sequences occurring by re-post actions performed by multiple users of the same social network. The method and system collects initial textual postings constrained by keywords and geo-locations of interest on a social network and constructs statistical models of user activities. Statistical models involve calculating interaction strengths between multiple users which is termed as edge

weight. Subsequently, statistical models also involve, calculating the postings, re-postings (replies) of multiple users, for over a period of time. Thereafter, an aggregated interaction of users is developed with the help of available statistical models.

Based on the aggregated interaction, the method and system selects a set of users along with their profiles whose social activity is found to be deviating from policies imposed on various social networks. Further, a set of posts and re-posts are also selected from the aggregated interaction which indicates certain data points such as , for example, emotions, sentiments, etc that are useful to the use...