/buttle-anon

Anonymisation work for DataKind DataDive, 2014

Primary LanguagePython

Named Entity Recognition (NER) & Cleansing

Masks people, organisations & locations in free text fields. Currently hardcoded for particular columns in my dataset, though this can be changed easily.

Usage

python nlp.py source_data.csv output_file.csv

Licensing

All my code is licensed under an MIT License. See NLTK documentation for details of its license.