[TIKA-1876] Integrate Natural Language Toolkit (NLTK) into Tika to perform Named Entity Recognition

• Type: New Feature
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.13
• Fix Version/s: 1.13
• Component/s: parser
• Assignee: Chris A. Mattmann
• Reporter: Manali Shah
• Created: 26/Feb/16 06:31
• Updated: 02/Mar/16 07:27
• Resolved: 02/Mar/16 05:47

Description

Hi all,

Apache Tika already performs Named Entity Recognition using Open NLP and Stanford Core NLP. Natural Language Toolkit is another open source python library and I believe it will be a great idea to have NLTK integrated along with Tika.
NLTK can extract NER as well as classify them. For this purpose I, along with Prof Chris Mattmann have published NLTKRest, a python pip/setuptools installable module that exposes NLTK as a REST service.

I have tested the working of Tika along with NLTKRest on my local repository and will soon submit a pull request.
Link to rest server: https://github.com/manalishah/NLTKRest