Skip to content

Graphusion

Extract knowledge graph triples from text with a pipeline

Visit official site

What Graphusion is

The tool is described as a pipeline that extracts knowledge graph triples from text through concept extraction, candidate extraction and fusion.

What you can do with it

  • Process text files from a set directory as input
  • Define relations with a JSON file
  • Run the whole pipeline from the command line

Getting started

  • Create and activate a new conda environment
  • Use pip to install the packages listed in requirements.txt
  • Prepare input text files and a relation JSON file
  • Use preprocess.ipynb to convert data into the required format

Pros & cons

✓ What we like

  • Automation across large text data sets is listed
  • Flexible configuration and data set adaptation are named

! What to watch out for

  • A JSON file is required to define relations
  • Input text files must sit in a chosen directory

FAQ

What are the pipeline steps?

Concept extraction, candidate triple extraction and triple fusion are listed.

What input does it need?

Text files in a directory and a JSON file defining relations are listed.

Last reviewed: 2026-11-15

More AI coding tools tools

View all →

How we review