kangjw1208/Tanzanian_Water_Wells_Status

Taking a look at water wells in Tanzania and looking to predict their statuses. # Tanzanian_Water_Wells_Status ## Overview This project uses classificational modeling to accurately predict the current state of water wells that are spread throughout the African nation of Tanzania. The purpose for the prediction model is to aid The Tanzanian Water Project organization on finding water wells that are not functioning for them to go in and repair as wll as to give indicators on why the well is currently non-functional. We'll be focusing our efforts on a data set provided by the Tanzanian Ministry of Water. This data set, before cleaning, had roughly 60k data points as well as over 30 features for us to use in an effort to predict the status. The steps we'll be taking so solve this problem will be: * Fully grasping the problem at hand in Tanzania * Explore and scrub our data * Use various classification machine learning models * Anaylyze results and hypertune our models for peak performance ## Business Understanding Currently in Tanzania, there is a massive lack for almost half of their population with access to source of safe drinking water or improved sanitation areas. Already posing an incredibly tough challenege, various organizations across the world have spent hundreds of millions of dollars to try and help tackle this problem from the early 2000s when Tanzania came under incredibly large scrutiny to fix water crisis. Being an issue back then when 73% of locals didn't have any access to drinkable water, we can see we've made decent strides in trying to tackle this problem with the introduction of water wells across the various districts in Tanzania. Unfortunately, due to many different problems such as geographic location of the well, extraction techniques, installation techniques, and the group that manages the well, we're beginning to see that many of these wells require some tender love and care. Our goal here is help the Tanzanian government, who is responsible for funding the majority of the wells across their country, to identify …