Inspiration
We have tried to build a co-relation between the market data on social media based on mentions and the sentiment of users about the stocks to predict wheather it is good for you to buy or not.
What it does
Webpage will display the top 10 stocks based on sentiment and their mentions and user will have an option to search for any stock and see it's investment score.
How we built it
Used NLP library "Vader" to find out sentiment of the users about particular post after fetching real time reddit posts. Also, completed analysis of volume and price metric and at the end combined everything, store it in the database. The website is developed with React and Node.js framework whereas some apis are developed with the use of flask.
Challenges we ran into
Combining everything and come up with single prediction metric.
Accomplishments that we're proud of
Able to complete the UI and ML stuff together.
What we learned
NLP library to calculate sentiments of users and also learned about some UI implementation stuff.
What's next for Influx - Investment Guide
Add filter to volume so that more-easily manipulable can be excluded Include dynamic functionality so that, if a stock that is not being tracked is mentioned, it is automatically added to the list of tracked stocks Alternatively : Generate the list using an API (ex: Finnhub) so that it always includes all available stocks Sort by categories (Volume, Mentions, etc.) Push/email notifications when an alert is met Search stats by stock Enhance UI
Log in or sign up for Devpost to join the conversation.