About:

We all know that coronavirus pandemic is a very critical situation for the whole world. In it's early stage I have tried to build algorithms that can forecast cumalitive confirmed cases as well as total fatalities region wise. The intention is to understand the future danger and help prepare the goverments as well as healthcare organizations to prepare well ahead of time. One of my models secured 6th position globally.

Dataset:

I am providing the link of datasets that I have used for these models:

https://www.kaggle.com/c/covid19-local-us-ca-forecasting-week-1/data

https://www.kaggle.com/c/covid19-global-forecasting-week-3/data

https://www.kaggle.com/sudalairajkumar/novel-corona-virus-2019-dataset

Algorithms and Libraries:

I have used regression algorithms such as Ridge, Lasso with Polynomial features, time series forecasting model-facebook prophet and tried other models as well such as Random Forest, XGBoost, LightGBM etc.