Doug Smith, Noah Yoshida, Cami Carballo
Special thanks also to UTA ClaimBuster fact-checking team
- Code is primarily python2.7, please alias/run appropriately in your environment.
- All data in REQUIREMENTS.txt
pip install -r REQUIREMENTS.txt
- spacy english module:
python -m spacy download en
- Configure your environment and install the requisite packages as explained in dependencies.
- Start the django server in mysite/
cd mysite
python manage.py runserver
- Server should start up and listen on a local port(8000).
fact JSON structure:
{
date: Date
sourceList: []{
author: String
publisher: String
date: Date
url: String
}
distilledFact: String
factStrings: []String
factHash: Hash
confidenceScore: Number
}
distilledFact: - String summarizing this fact based on common interpretation of all fact strings
factStrings: - Actual quotes of all strings that reduce to this fact
factHash: - Hashed representation of what the fact concerns (Subjects x Sentiment)
confidenceScore: - Numeric representation of how often this fact has been presented/verified
BBC
Top Stories: http://feeds.bbci.co.uk/news/rss.xml
Politics: http://feeds.bbci.co.uk/news/politics/rss.xml
NYT
Home Page: http://rss.nytimes.com/services/xml/rss/nyt/HomePage.xml
US: http://rss.nytimes.com/services/xml/rss/nyt/US.xml
Politics: http://rss.nytimes.com/services/xml/rss/nyt/Politics.xml