Leviafunc(legal_acts):
    return work_of_Leviathan

NLP of legal texts

Analysis of agreements between governments

Key words & key phrases extraction with TF-IDF and N-gramms
NER for DATES with (Natasha (rule-based lib for Russian language). Sequence model, implemented in AnaGo and NER by DeepMIPT both have lower accuracy for this type of documents.
Dictionary method with morphological analysis for finding ORGANIZATIONS and COUNTRIES

use as a module

$ pip install leviafunc

use cases are described in example.py

or run as a script

$ git clone https://github.com/alissiawells/Leviafunc.git
$ cd Leviafunc
$ mkvirtualenv Leviafunc
$ pip install -r requirements.txt
$ python leviafunc.py input.txt [-f / output.txt] [-c yourcorpus.txt]

run without [additional options] for console output with initial corpus ('corpus.txt')

tests:

python setup.py test

for experiments, it is more conveient to use jupyter notebook

$ jupyter notebook

open on localhost 'Analysis.ipynb'