Mining graphs with constraints on symmetry and diameter

Abstract

The area of graph mining bears great importance when dealing with semi-structured data such as XML, text and chemical and genetic data. One of the main challenges of this field is that out of many resulting frequent subgraphs it is hard to find interesting ones. We propose a novel algorithm that finds subgraphs of limited diameter and high symmetry. These subgraphs represent the more structurally interesting patterns in the database. Our approach also allows to decrease processing time drastically by employing the tree decomposition structure of database graphs during the discovery process. © 2010 Springer-Verlag.