Previous research has shown a connection between homelessness, mental health, and drug addiction. Mental illness can make it more difficult to keep up with daily activities which can make it more difficult to keep housing. The inability to gain access to proper mental health care or experiencing homelessness can make it more likely for people to turn to illicit drugs which in turn can make housing more difficult to find. These are complicated issues that form a muti-way relationship (Tarr, 2018). Homelessness is a socioeconomic issue that every state in the United States faces. It can be agreed upon that homelessness is a multifaceted issue with many causes.

In this research I sought to determine if I could use a prediction model to determine a states percentage of its population that is experiencing homelessness. To accompish this I needed to answer what factors affect a state’s homeless population. I chose to do my project looking at the population of people experiencing homelessness in America in relation to drug use and mental health. I also wanted to pose the question, could addressing these factors help tackle the problem of homelessness.

The contents of this repository includes my paper, slides, video presentation, code, and data set. This is the ccollection of work that I put together for my data science practicum II course. I believe that looking at these issues with prediction models will allow us to create plans and projects to move our commmunities forward.

NOTE: The data set that I gathered for this project is limited and does not include enough points to create a reliable neural network however I want to keep that as a part of the project to show the code and the possibility of it being useful were I to have city data which would allow for many more data points.