carimura / jlike

initial proto of sentiment analysis.

JLove - Sentiment Analysis in Java

JavaNLP API Docs

Test with: 1158420985915117579

TODO:

• simple html template would be great
• create bearer token if doesn't exist
• Influencer score of retweeters
• add Fibers
• add jlink and Docker?
• sentiments seem low...
• train model -- feed info back to model.
• to include models or not to include them, probably not. too big. maybe download at Docker build time and utilize its caching

Build

mvn package

Setup

❯ curl -u "$TWITTER_CONS_KEY:$TWITTER_CONS_SECRET" \
  --data 'grant_type=client_credentials' \
  'https://api.twitter.com/oauth2/token'

❯ export TWITTER_BEARER=<result from above>

Run

java --enable-preview -jar target/jlike.jar

--> http://localhost:8080/
--> http://localhost:8080/twitter/<username>