RHOG is a Java library to operate with directed labeled graphs (DLGs), that includes:

• Import/export to formats such as DOT, GML, GraphML, TGF, and more.
• Operations on DLGs:
  • basic vertex/edge addition/elimination
  • graph type checking (trees, lattices, DAGs, connected graphs, etc.)
  • subsumption checks
  • anti-unification
  • disintegration
• refinement operators: a large set of refinement operators are included in the library, supporting graphs with flat labels, or graphs whose labels are part of an ontology, etc.
• basic machine learning algorithms based on the refinement operators (distance and similarity measures for use in nearest neighbor classifiers)
• graph visualization tools for easy debugging

An article describing all the theory behind the functionality presented in this library can be found here.

Project supported by NSF grant IIS-1551338.