What is Polarity?
Do you want to know if your tweets come across as cheery or grumpy? Polarity uses machine learning to read your latest tweets and judge their positivity. It also gives you your most-used words, and tells you whether they are positive or negative. It's a great and quick way to get a view of the emotional content of your tweets, or anyone else's.
How it Works
Polarity uses the Support Vector Machine (SVM) algorithm built into the scikit-learn library. We trained it on pre-classified data from http://help.sentiment140.com/for-students/ . However, the flexibility of the machine learning approach is that it can easily be trained on any data set.
How to use it
Check it out from the github repository, then run predictor.py. It should print out:
- Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)
Copy the IP address in to your browser and add test to the end, for example:
Then put in your twitter account!
Credits
Jacqueline Chen - Javascript, HTML, design
Mark Creamer - Twitter API, javascript
Jon San Miguel - Machine learning, language processing, backend
Built With
- d3.js
- flask
- javascript
- ntlk
- python
- scikit-learn
- tweepy
Log in or sign up for Devpost to join the conversation.