Modern knowledge graphs (KGs) are built using a combination of structured data, crowd sourced contributions, and the output of information extraction from documents, images and video. Their content is rich and noisy, challenging the rigidity of traditional ontologies and logic based-reasoning. Modern KG-reasoning methods emphasize statistical reasoning based on deep learning. The objective of KGTK is to build the "Scikit-learn" of KGs, a comprehensive library of tools and methods to make it easy to create, integrate, denoise, reason and query KGs to build interesting applications that leverage vast amounts of knowledge. KGTK is designed to support reasoning pipelines composed of operators to import, filter, transform, abstract and reason with KGs. A simple edge-list representation of KGs using CSV files enables easy incorporation of high performance utilities and packages written in different languages. The expectation is that application developers will find it easy to compose and experiment with different KG pipelines in Jupyter notebooks, similarly to how data scientists today can rapidly build data driven models using Scikit learn.
Hands-on materials: https://github.com/usc-isi-i2/kgtk-notebooks/
Colab Notebooks: https://github.com/usc-isi-i2/kgtk-notebooks#running-the-notebooks-in-google-colab
KGTK documentation: https://kgtk.readthedocs.io/
Similarity GUI: https://kgtk.isi.edu/similarity/
KGTK Search: https://kgtk.isi.edu/search/
KGTK Browser: https://kgtk.isi.edu/iswc/browser/Q2685
Resource paper (ESWC'20): https://arxiv.org/pdf/2006.00088.pdf
KGTK on GitHub: https://github.com/usc-isi-i2/kgtk/