Inspiration

** We have been inspired by both of finance and AI&ML topics. We wanted to create a project which includes those topics. **

What it does

** It successfully estimates the changes on the different stocks for a given period of a time and a specific company.**

How we built it

** We preprocessed the data that we took from the Yahoo API and we did some feature engineering to increase the accuracy of the learning process. We fed the ML algorithm by this preprocessed data. After that, we created a web application using streamlit library. **

Challenges we ran into

** The first obstacle we have encountered was lack of features of the data. So we did some feature engineering in order to increase the accuracy. The second thing was to overcome the overfitting problem. In order to fix it, we dropped out 20 percent of the neurons during the training process. The third thing was that it was quite hard to increase the accuracy for lengthy inputs, which refers to longer time periods.**

Accomplishments that we're proud of

** We successfully taught the LSTM neural network and got accurate predictions.**

What we learned

** We learned how to manipulate and organite the data. Also we gained a hands-on experience in Deep Learning. **

What's next for Stock Investment Advisor

** We aim to improve its general software design. In addition to this, we want to optimize the ML algorithm and fed it with larger data sets.**

Built With

Share this project:

Updates