The demo page is at http://big-data.ee.ucla.edu/demo/
This demo page is for entity relationship extraction. The user may input some texts in the textarea and get the entities and relationships in the texts by simply clicking the submit button. If any relationships found in the text, three tables will show up and display the extracted relationships, ranking of the extractions, and ranking of the entities respectively. To make the result vivid, a graph is developed to display the entities as nodes and relationships as links.
The demo page for entity relationship extraction is developed with Flask.
The graph is implemented with D3.v3.js.
- practnlptools can be downloaded from https://github.com/biplab-iitb/practNLPTools
python setup.py install
- nxpd:
sudo pip install nxpd
- nltk:
sudo pip install nltk
python -m nltk.downloader all
- Flask 0.10.1
sudo pip install Flask
sudo pip install flask_wtf
- For local test:
python run.py
See the webpage at http://localhost:5000
- To test on a server:
python serve.py start | stop | restart
index page http://big-data.ee.ucla.edu/demo/:
