Metadata-Version: 2.1
Name: person-extractor
Version: 4.0.0
Summary: Extract people from text
Home-page: https://github.com/Mak4Lab/person-extractor
Author: Daniel J. Dufour, Victoria D. Mak
Author-email: daniel@mak4lab.com, victoria@mak4lab.com
License: UNKNOWN
Download-URL: https://github.com/Mak4Lab/person-extractor/tarball/download
Description: # person-extractor
        Work in Progress: Identify People's Names in Text
        
        # usage
        You initialize a PersonExtractor with a path to a CSV of names with each column a language.
        You can create a csv through [Wikinames](https://github.com/mak4lab/wikinames).
        
        ```python
        from person_extractor import PersonExtractor
        
        text = "Але дістатися на роботу працівникам цих бізнесів, якщо у них немає власного автомобіля або грошей на таксі чи корпоративну розвозку, стане справжньою проблемою, прогнозує політолог Микола Давидюк."
        
        extractor = PersonExtractor(data="names.csv")
        
        people = extractor.extract(text)
        ```
        
        extract returns a list of objects:
        ```python
            [
                {
                    'start': 336,
                    'end': 343,
                    'text': 'Давидюк',
                    'spellings': {
                        'en': 'Davidyuk',
                        'uk': 'Давидюк'
                    }
                }
            ]
        ```
        
        # test
        To test the package run:
        ```
        python -m unittest person_extractor.test
        ```
        
        # contact
        Post an issue at https://github.com/Mak4Lab/person-extractor/issues or email the package authors at daniel@mak4lab.com and victoria@mak4lab.com 
        
Keywords: entity extraction,person
Platform: UNKNOWN
Classifier: Development Status :: 3 - Alpha
Classifier: Programming Language :: Python :: 3
Classifier: License :: CC0 1.0 Universal (CC0 1.0) Public Domain Dedication
Classifier: Operating System :: OS Independent
Description-Content-Type: text/markdown
