Discovering Interesting Subgraphs in Social Media Networks

TL;DR

This paper introduces an algorithm for automatically discovering interesting subgraphs in social media networks by comparing subgraph structures and content to a user-defined background graph, enhancing social media data analysis.

Contribution

It presents a novel method combining group-by operations and subjective interestingness to identify structurally and content-wise distinct subgraphs in heterogeneous social media graphs.

Findings

Effective discovery of interesting subgraphs demonstrated on socio-political data

Algorithm identifies subgraphs that differ significantly from background graphs

Method enhances understanding of social media network structures

Abstract

Social media data are often modeled as heterogeneous graphs with multiple types of nodes and edges. We present a discovery algorithm that first chooses a "background" graph based on a user's analytical interest and then automatically discovers subgraphs that are structurally and content-wise distinctly different from the background graph. The technique combines the notion of a \texttt{group-by} operation on a graph and the notion of subjective interestingness, resulting in an automated discovery of interesting subgraphs. Our experiments on a socio-political database show the effectiveness of our technique.