Triple S

We built this because we thought it'd be really cool to analyze music – not just for its tone but also the inherent lyrics. Are dancing songs necessarily those with happy (positive) lyrics? How often do high energy songs have negative lyrics? Are songs with more acoustic more likely to show up on a top 100 list? IS THE MICROSOFT API BETTER FOR ANALYZING LYRICS, OR IS IT GOOGLE? Find out today at 12PM.

Name of the project: Triple S

Elevator pitch: Symphonies' Sentiment Scrutinizer. An app to give weekly trends of music that is considered because based on the sentiments it conveys, ambiguity in emotions expressed, fast and loud but sad songs, low energy but danceable songs...

Screenshot: Please make a screenshot of your project and insert it here (use markdown)

Type of the project: web app, mobile app, data visualization? Web App, data visualization.

Target group: Who should use your product? Everyone! Anyone who listens to music. Anyone who is interested in curious data. Anyone who wants to think about why Microsoft and Google disagree about sentiment songs convey – i.e. everyone

Challenges & solutions: What was the most difficult part of the project? What problems did you encounter? How did you solve them? Thinking about the different ways to look at data, using Spotify, Google and Microsoft APIs to analyze songs both from a musical perspective and from a lyrical perspective and drawing inferences based on the data is what we did. In the way, figuring out how to use beautiful graphs to display this data for Javascript was annoying, and learning Flask on the fly was almost magical.

Scalability: How can the project be further developed? Users' data can be analyzed, and users can find out about their music listening trends. Do they often listen to sad songs? Or happy songs? Maybe in certain weeks they listen to slower songs. All this data can help predict user's moods effectively and contain insightful information for the user at hand.

Code: Python {Flask}, Javascript

Technology: Which frameworks and solutions did you use? We used the Spotify API, Microsoft Text Analytics API, Google's Natural Language Processing (Sentiment) API, Lyrics fetching API, and finally hosted our Web App using Google's Cloud services.

Link: triple-s-156419.appspot-preview.com

Source code (optional): Please insert the link to your project's GitHub repository. https://github.com/raghavanand98/Sentiment-Experiments

Video: https://drive.google.com/drive/folders/0B722AQunIepzRklyNEtDZG1aelU?usp=sharing

Team: Mudit Gupta and Raghav Anand.

Team members, their skills & Twitter handles: -Raghav Anand (full stack) -Mudit Gupta (full stack)