Social Media Monitoring via Sentiment Analysis
# clone this repository
git clone
# cd into the project directory
cd kaliVisuals
# install frontend dependencies
cd frontend
npm install
# install functions dependencies
cd functions
npm install
# install backend dependencies (optional)
cd functions
npm install
# go to the functions directory
cd functions
# create new file
touch twitterApiKeys.js
# update the following after obtaining your keys
module.exports = {
consumer_key: 'YOUR_CONSUMER_KEY',
consumer_secret: 'YOUR_CONSUMER_SECRET',
access_token:'YOUR_ACCESS_TOKEN',
access_token_secret: 'YOUR_TOKEN_SECRET',
timeout_ms: 60*1000,
strictSSL:true,
}
-
React - The entire client side is built using React and styled using Semantic-UI-React
-
Firebase - Firebase Auth, Firebase Realtime Database and Cloud Functions are powering the entire backend of this web application
-
Sentiment - AFINN-based sentiment analysis for Node.js
-
Twitter-API - Parsing hundreds of tweets/call/day and aggregating sentiment score. Working in conjunction with cloud functions running Node, to parse tweets and calculate sentiment
-
Cron-Job - Daily scheduled execution of fetching and analyzing sentiment data before displaying the data on the dashboard. looking to move the scheduling of tasks to google compute engine
-
Warning -- upon user deletion of a monitored hashtag, a warning is emitted due to chart unmounting incorrectly. Most probably have to do with the DB listeners still attached to chart after deletion
-
When creating a new user, the avatar image doesn't always load; it requires a refresh of the page.
- Change user avatar
- More API sources
- Search Functionality