This API was made to solve the issue of sentiment analysis for emotion detecion in lyrical corpi.
This API serves up 2 recommendation engines that do the following:
-
Song Emotion Endpoint
- Given a string as an input, this makes use of a multiclass ElasticNet regressor to determine the following emotions for a given song:
- Anger
- Fear
- Joy
- Sadness
- Surprise
- Given a string as an input, this makes use of a multiclass ElasticNet regressor to determine the following emotions for a given song:
-
Similar Songs
-
Given a JSON of emotions that a given song could contain, this uses the K-Nearest Neighbors algorithm to retrieve songs that are nearest in distance to the emotional values submitted. This returns:
- Artist Name
- Song title
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.
To get started with Using this API please refer to : https://github.com/Novandev/gn_api/wiki
This package requires Python 3 to be installed
- Clone the repo
git clone https://github.com/Novandev/gn_api.git
- Install dependencies from project root folder (Please utilize virtualenv, or pipenv)
pip install -r requirements.txt
- Run the project
EXPORT FLASK_RUN=app.py
or
python3 app.py
The server by default will run on port 5000
FOr post examples refer to: https://github.com/Novandev/gn_api/wiki/All-API-Endpoints
Explain how to run the automated tests for this system
Add additional notes about how to deploy this on a live system
- Flask - The Python backend Framework
- SK-Learn - Machine Learning Framework
- Pandas - DataFrame and Analysis Framework
- AWS ElasticSearch - Document storage and retrieval
- IndicoIo - Sentiment Analysis API
Please read CONTRIBUTING.md for details on our code of conduct, and the process for submitting pull requests to us.
- ** Novan Adams ** - * Data Scientist and App Developer * - Novandev
See also the list of contributors who participated in this project.
This project is licensed under the MIT License - see the LICENSE.md file for details
- [https://www.gifnote.com/] - Gifnote