Home

Neji

Neji is a flexible, easy and powerful framework for faster biomedical information extraction.

It allows you to

Build your processing pipeline

1. Read documents
2. Process
3. Concept tree
4. Store

How to use it?

• CLI
• SDK

Resources

• GDep optimised for the biomedical domain
• Gene/protein NER model trained on GENETAG
• Chemical NER model trained on BCV

License

Attribution, share alike and non-commercial use (CC BY-NC-SA 3.0).

What you can do with Neji?

• Concept recognition:
  • Dictionary-based
  • Machine learning-based
  • Rule-based
• Train machine learning models for NER (Named Entity Recognition)
• Linguistic parsing:
  • Sentence splitting
  • Tokenisation
  • Lemmatisation
  • Chunking
  • Dependency parsing

Installation

Use

Training new models

Resources

external tools

models

dictionaries

Input and output formats

SDK

annotation

training