Automatic Identification of Abstract Online Groups

Battelle Number: 17007 | N/A

Technology Overview

Using algorithms to find online groups via social media sources, PNNL scientists have developed a method to automatically identify online groups that exhibit shared behaviors, interests, and/or characteristics. Our method uses both explicitly connected groups and ‘abstract’ groups, in which the connection between individuals is interest and behavior rather than explicit links.  These variables do a good job of identifying groups, placing members within a group, and helping determine the appropriate granularity for group boundaries. The group footprint can then be used to identify differences between the online groups.

Using structured experiments it can be shown that cultural differences do exist in both off-line and on-line communities. This is evident in the culturally sensitive adoption, use, and behavior on the Internet.  However, the first and possibly toughest step is to identify the online communities and then identify the quantifiable dimensions of possible cultural differences within the groups. Using the hypothesis that quantifiable cultural differences exist within online groups, the characteristic footprint used to identify online groups also can be used to demonstrate cultural differences between individuals within a group and between common cultures across different groups.



  • Demonstrates cultural differences between individuals in online groups.
  • Determines what kinds of interaction there are between levels and types of groups
  • Provides significant information for use in areas ranging from targeted marketing to intelligence applications.


Available for licensing in all fields


Abstract groups; clustering; content-based; footprint; anomaly detection; cultural differences; automatic identification; online groups

IP files


CY-Enterprise Cybersecurity

Market Sectors