Assignment Task 1

Assignment Task 2

Application Task 3

Pre-requisites:

• Steps to be followed form Task #1

1. Set up a web application on Heroku or Digital Ocean Droplet with a user interface where we can input a named-entity and get the comparison between the news and twitter sentiments as an output.

   • The application is a nodejs, expressjs api that handles the data and sends it to the front end in JSON format. API fetches real time tweets using Twitter Search Api and the News Everything API. For sentiment analysis, Aylien's Sentiment Ananlysis api is used.
   • The front end is written in React and present in the front-end directory. I have used the create-react-app package to start with boilerplate code.

   API doc:

   Verb	Route	Description
   POST	/tweet/:keyword	to fetch top 5 tweets and return their sentiments.
   POST	/news/:keyword	to fetch top 5 news articles and return their sentiments.

   Front End Snapshots:

   By default it shows result based on the keyword: "technology"

   The result changes when you enter a keyword: "politics"

Deployments

• The api and front end are deployed on different servers and data is transferred/received using AJAX requests.
• API - is served using bin/www and pm2 for monitoring and logging.
• Front End - a production build of react is served using nginx server.

2. Put all your code, along with the MongoDB collection, in a GitHub repository and share the link with us. Also, maintain a README.md explaining your codebase and the approach you followed.

Help links:

• Stack Overflow
• The Python Graph Gallery
• Mongodb Documentation
• Mongoose ODM
• Pymongo Documentation
• Chloropeth Maps