/Generate-and-Rank-ConNorm

Generate-and-Rank Framework with Semantic Type Regularization for Biomedical Concept Normalization

Primary LanguagePython

Generate-and-Rank-ConNorm

Generate-and-Rank Framework with Semantic Type Regularization for Biomedical Concept Normalization

Setup

  • We recommend Python 3.7 or higher. The code does not work with Python 2.7.
  • BERT-based multi-class classifier and BERT-based listwise classifier are implemented with PyTorch (at least 1.3.0) using transformers v3.0.2.
  • Lucene-based dictionary look-up system is developed using the lucene-8.1.1.

Getting Started

Candidate Generator

Candidate ranker