Inspiration
What it does
Gets sentiment of a keyword or phrase based on twitter status's. This sentiment is then graphed for a week of data so that the user can see the trends of the specified keyword on twitter. Things like 'I Like' and "I love" are weighted toward positive sentiments whereas phrases like "I hate" or "I am sad" are weighted toward negative sentiments. These weights are then averaged between an assortment of related tweets to find an average sentiment for the keyword.
How I built it
We used a php back-end and a javascript front-end to handle some animations.
Challenges I ran into
HP API is very, very, very slow. I repeat, very, very, very slow...
Accomplishments that I'm proud of
Connecting to twitter and querying tweets based on different options.
What I learned
For a less general-purposed application, it would be more efficient to write your own sentiment analyzer
What's next for Wild Tweets
We will cache searches done by users to increase speed and efficiency. We will also work on having different data visualization techniques and possibly allow for more complex queries by the user.
Log in or sign up for Devpost to join the conversation.