Millions of dollars are invested in algorithm trading and we needed to develop a open source model which could predict stock market pricing from the viewpoint of long term as well as short term trading.
What it does
Predict the opening price, closing price for stocks in the future. Takes help of sentient analysis of the new feed to tap on the mood of the market. Stock market is an largely based on the actions and decisions of every individual participating in it and every user's action has a n impact on it.
How I built it
Trained Recurrent Neural network using tensorFlow in the backend, sentiment analysis of news feed trained and saved in a database created using the Hasura api.The front end being a web app provides an interactive platform for the user to utilize the pre-trained models for visualizing the predictions.
Challenges I ran into
tensor flow training of the data
Accomplishments that I'm proud of
integrating the trained models, database, user interface for a smooth interaction
What I learned
team work, full stack integration
What's next for StockAdvisors
analyze weather to predict the trends of the commodity trading market