A full-stack application that retrieves tweets via Twitter API and analyzes them using natural language processing (NLP) to determine positive, negative, or neutral sentiment for selected cryptocurrencies.
Visual demonstration of DigiCurr Sentiment in use:
Web Deployments: DigiCurr Sentiment on Heroku
In order to successfully run DigiCurr Sentiment after downloading this repo, you will need to obtain API credentials from Twitter. To obtain your API credentials, visit https://developer.twitter.com/en/apply-for-access.
- Download this repo
- Obtain your Twitter API credentials
- Use the ".env.example" file as a template to create your .env file containing your API credentials
- Open a Node terminal and change the directory to your folder containing the contents of this repo
- Run "npm install" to install the dependencies
- Run "node server.js" to launch on http://localhost:5000 and open it in your browser
This project is licensed under the AGPL-3.0-only license - see the LICENSE.md file for details.
This is my fourth independent project since I began learning how to program in early 2018.
After finishing my last project, I had a few things on my mind about future project ideas, existing interests that I wanted to incorporate into those projects, and cutting-edge areas I'd like to learn more about. I decided to bring my existing interest in cryptocurrencies together with my desire to learn more about artificial intelligence (AI).
The project name, DigiCurr Sentiment, describes what the project is about. "DigiCurr" is a portmanteau of "digital" and "currency," since cryptocurrencies are a major type of digital currency, and "Sentiment" points to what the project is analyzing and measuring.