mughetto / onto_merger Goto Github PK

Paper | Documentation | External Resources

OntoMerger is an ontology alignment library for deduplicating knowledge graph nodes,(i.e. ontology concepts) that represent the same domain, e.g. diseases, and connecting them to form a single directed acyclic hierarchical graph (DAG) (i.e. an ontology class hierarchy). The library implements a pipeline that takes nodes, mappings and (disconnected) hierarchies as input and produces node merges and a connected hierarchy. It also provides analysis and data testing for fine tuning the inputs in order to further reduce duplication, as well as to increase connectivity.

Citing

If you find OntoMerger useful in your work or research, please consider adding the following citation:

@misc{ontomerger,
  doi = {10.48550/ARXIV.2206.02238},
  author = {Geleta, David and Nikolov, Andriy and ODonoghue, Mark and Rozemberczki, Benedek and Gogleva, Anna and Tamma, Valentina and Payne, Terry R.},
  title = {OntoMerger: An Ontology Integration Library for Deduplicating and Connecting Knowledge Graph Nodes},
  publisher = {arXiv},
  year = {2022},
}

Getting Started

from onto_merger.pipeline import Pipeline

# initialise the pipeline
pipeline = Pipeline(project_folder_path="../path/to/project")

# run the process
pipeline.run_alignment_and_connection_process()

# view results in "../path/to/project/output/report/index.html"

Running tests

$ tox -e py

License

• Apache 2.0 License

Credit

The Onto Merger logo is based on:

• Galguna Font
• Noun Project Icons