Sentwee

• 

  The graph, showing sentiment vs time

• 

  On hovering on a point, the tweet message appears in the tooltip

• 

Inspiration

Pretty graphs, social media

What it does

Takes a realtime feed of a certain searchterm from twitter, puts it through the google cloud api to find the sentiment (happiness) and then displays it in a web app

How we built it

In Python, Tweepy collects the tweets, which are passed to Google Cloud to find the sentiment. These are sent over a socket to the Flask webserver, which sends them over a continuous event stream to the front end, where CanvasJS plots them. As such, everything is entirely real time

Challenges we ran into

• Forgetting/not knowing python
• Getting the twitter and server process to talk to each other - streams going all over the place
• API rate limits

Accomplishments that we're proud of

• IT WORKS
• Everything being in real time

What we learned

• Python
• JSON
• Using APIs
• Realtime streams
• Who Panda is

What's next for Sentwee

• Better UI
• Selectable search term on website

Built With

• python
• tweepy
• google-cloud
• flask
• canvas-js