Stock-Price-Forecasting-Using-Artificial-Intelligence

New approach for stock market prediction using Artificial Intelligence

Techniques used

Bidirectional LSTM on historical data

Sentiment analysis with news and twitter feeds

Django + Angular JS combination for web app

Requirements

Python packages

LAMP installed

Django

celery

keras

tensorflow

pandas

numpy

JS packages

Jquery

Angular JS

Chart JS

How to run

first terminal

bash git clone git@github.com:koriavinash1/FIN_ishers.git cd Stock-Price-Forecasting-Using-Artificial-Intelligence/StockNest python manage.py makemigrations python manage.py migrate python manage.py runserver

second ternimal

cd Stock-Price-Forecasting-Using-Artificial-Intelligence/StockNest celery -A StockNest worker -l info

First steps

• data collection: Once server is on, navigate to localhost:8000/stocksadmin download all the required data

• train model from same stocksadmin page

• once model is trained, you can use product from localhost:8000/index paage

DL model information

• DL model information can be found in FINishers/StockNest/stockbackend

• Django restapis can be found from all apps/apis.py script

Predictions

Observed RMSE < 0.05 on test data

Contact

• Avinash Kori (avinashgkori@smail.iitm.ac.in)

If any comments or information required, pull requests/issues are Welcomed....

Thankyou