Inspiration
We wanted to do something which was known to be a challenging problem, and solve it with the cutting edge tools provided for us. We decided to tackle the difficult problem of computational finance and machine learning based stochastic forecasting.
What it does
A UI which assists users in investing decision by providing stock forecast based on big data mined from Twitter Feeds, New York Times, and Yahoo Finance.
How we built it
We used Azure ML to build an Artificial Neural Network regression algorithm. The web app was built with python, flask, html5, jquery, and chart.js
Challenges we ran into
Azure ML Training and API challenges were encountered. Azure ML training took a good 8 hours on complex models and the API was difficult to work with at times.
Accomplishments that we're proud of
The sheer accuracy of our machine learning algorithm and the beautifully simple design.
What we learned
The stochastic nature of the stock market is difficult to model with machine learning based regression algorithms.
What's next for StockBrief
We would like to build a hedge fund with our new algorithm.
Log in or sign up for Devpost to join the conversation.