Accession Number : AD1051096


Title :   Content-Based Covert Group Detection in Social Networks


Descriptive Note : Technical Report,01 Oct 2013,31 May 2017


Corporate Author : Carnegie Mellon University Pittsburgh United States


Personal Author(s) : Sycara, Katia


Full Text : https://apps.dtic.mil/dtic/tr/fulltext/u2/1051096.pdf


Report Date : 06 Sep 2017


Pagination or Media Count : 15


Abstract : Along with the benefits of providing ways for people to connect and socialize on-line, social media provides opportunities for various groups to spread messages to try to spread influence and shape opinions according to interests of the group. Community detection in social media is a challenging problem. A community can be defined as a group of users that (1) interact with each other more frequently than with those outside the group and (2) are more similar to each other that to those outside the group. Research on detection of groups that disseminate content to push their own agenda faces many challenges, due among many others, to the difficulties of modeling and detecting cues for strategic messages of groups. To help close this research gap, the overall goal of this project is to develop techniques and algorithms for detection of goal-driven covert groups that spread strategic information.


Descriptors :   online communications , natural language computing , social networks , machine learning , computational linguistics , automated text summarization , classification , accuracy , hierarchies , data mining


Subject Categories : Information Science


Distribution Statement : APPROVED FOR PUBLIC RELEASE