[OFBIZ-7954] Add NLP function to parse text content

• Type: Improvement
• Status: Open
• Priority: Minor
• Resolution: Unresolved
• Affects Version/s: Trunk, Upcoming Branch
• Component/s: solr
• Labels: NER NLP
• Assignee: Shi Jinghai
• Reporter: Shi Jinghai
• Created: 15/Aug/16 06:20
• Updated: 11/Jan/22 08:48

Description

NLP(Natural Language Processing) is an amazing tech can help us to improve our beloved OFBiz.

NLP can be used to train your own model to parse product attributes and understand what it is, before importing the product, or used as an online CRM text robot to answer simple questions.

We will try to add a new nlp component, and we hope we can get some help from the community.

AFAIK, two apache projects are working on NLP:
http://tika.apache.org/
http://opennlp.apache.org/

The Stanford NLP is used in tika:
http://nlp.stanford.edu/

While implementing this component, we found an ebay paper on this topic is very useful:
http://aclweb.org/anthology/D/D11/D11-1144.pdf

Kind Regards,