An open API service indexing awesome lists of open source software.

https://github.com/tigregotico/pronomial

pronomial postag/word_gender based coreference solver
https://github.com/tigregotico/pronomial

coreference coreference-resolution coreference-solver natural-language natural-language-processing natural-language-understanding nlp pronouns

Last synced: 9 months ago
JSON representation

pronomial postag/word_gender based coreference solver

Awesome Lists containing this project

README

          

# Pronomial

simple and fast coreference solver, uses pos tags, a word gender classifier
and lists of pronouns

Supported languages: English, Portuguese, Spanish, Catalan

# Install

```bash
pip install pronomial
```

# Usage

```python
from pronomial import replace_corefs

replace_corefs("London has been a major settlement for two millennia. "
"It was founded by the Romans, who named it Londinium.",
lang="en")
"""
"London has been a major settlement for two millennia . "
"London was founded by the Romans , Romans named London Londinium ."
"""
```

## About

Pronomial will work fine for short sentences and should be safe to use in
the context of simple things, but it tends to fail horribly in complex sentences.

It you want a proper coreference resolution library I suggest you check out
[neuralcoref](https://github.com/huggingface/neuralcoref) or [coreferee](https://github.com/msg-systems/coreferee)

Pronomial should not be used in production most of the
time, it is intended as baseline and an experiment on how well this
task can be solved using only simple heuristics.