Deliver a Google Chrome Extension that analyzes the entities (individuals, organizations, countries, etc.) contained within a news article and categorizes the most important entities as either being a hero, villain, or victim.
pip3 install newspaper3k
Mac/Unix Enviornment:
pip3 install nltk
python3 -m nltk.downloader punkt averaged_perceptron_tagger maxent_ne_chunker words wordnet
Windows: http://pypi.python.org/pypi/nltk
pip3 install textblob
pip3 install spacy
python3 -m spacy download xx
pip3 install flask
Open the Extension Management page by navigating to chrome://extensions.
Click the LOAD UNPACKED button and select the extension directory.
python3 flask_app.py
Open the a news website on Google Chrome and click the extension icon next to right of the address bar.
- Newspaper - scrape articles from news sites
- NLTK - recognize entities and compute word similarity
- TextBlob - analyze sentiment of words
- Spacy - recognize active/passive sentences
- Flask - host a local server to run code
- Tianna Avery - tiannaavery
- Quinn Mayville - mayvilleq
- Yingying Wang - yingyingww
- Megan Zhao - meganzhao
- Inspired by the paper: Who is the Hero, the Villain, and the Victim?: Detection of Roles in News Articles using Natural Language Techniques (https://dl.acm.org/citation.cfm?id=3172993)
- Basis of merging algorithm comes from the NU Infolab News Context Project
- Advised by Professor David Musicant of Carleton College
- September 2018 - March 2019