Description / Project Intention
During times of crisis, social media has been the main platform for airing opinions, good and bad, practical and absurd. These can make for a read that’s informational, entertaining, or a combination of both. Our Twitter mood analyzer will take a user-provided tweet thread and using IBM’s Tone Analyzer, display the overall emotion of each post and its series of responses as a map of colorful points. The user can click on these points to see a more detailed breakdown of how the Tone Analyzer understood the nuances of a particular response. The analyzed tweet will also be embedded on this page for direct comparison.
Inspiration
COVID-19 was partially an inspiration, but we also wanted something that could be used as a light search tool for curiosity during large (or small) scale events that generate discussion. Our web app allows others to hone in on specific Tweets conveying a specific emotion in a conversation. Instead of reading through comments one by one, they can start with a Tweet that sparks interest, whether they're looking for positive content for comfort or angry statements for informational purposes.
What We Learned
Learning New Concepts:
- Data Models: many-to-many, one-to-many, many-to-one (+ unidirectional/bidirectional variants)
Learning New Frameworks:
- ORMs/databases
- Bootstrap
- Flask
Best Practices:
- Consistency in design: visually and structurally
- Organization
- Distribution of tasks (backend vs. frontend distinction)
- Directory structuring & packaging
- Proper use of version control (Git)
Communication:
- Avoiding merge conflicts
- Switching / handing off tasks
- Checking in with each other
- Discussing major changes with the entire team before committing


Log in or sign up for Devpost to join the conversation.