The social problem that I am implementing in my project is about our recent health crisis in the US. The Spanish Flu occurred more than a hundred years ago and due to its inefficient technology, the virus ended up killing around 50 million people globally. However, we should take advantage of the current science and modernized technology that we have developed over the past 100 years to ensure that we do not repeat our mistakes.

As a resident living in the United States, I am really worried about the future of our country and I wonder if the virus will expand or contract in the future. By using past data to predict the future, this project will also inform the residents of the United States how the country will be if we continue this path. This architecture will also predict the number of deaths if more people decide to wear masks or less people. This model will not only use GitHub data, but also utilizes data from John Hopkins in order to determine the future number of deaths. Thus, as this architecture will predict the future, the residents will begin to understand the severity of the situation and will begin to follow the guidelines that CDC has stated in their website such as social distancing or wearing masks.

In this architecture, I used auto-regressive models(type of linear regression model) to analyze historical time series data and to predict the number of deaths a week ahead. I picked 10 states in the United States and used github data from each state in order to predict the future.

My code: