kenwelsh/Tanzania_Water_Pump

Machine learning model to predict one of three classes for water pumps in Tanzania # Tanzania_Water_Pump Machine learning model to predict one of three classes for water pumps in Tanzania ### Team Members + Kuntal Deshpande + Matt Glibota + Ken Welsh + Nabanita Gupta + Allen Shih ### Project Summary This project is to build a machine learning model to classify water pumps in Tanzania as either functional, functional – needs repair, or non-functional. We chose this project because access to clean water is a significant issue faced by many communities around the world. Around the world, 785 million people lack access to safe water. Access to safe water helps empower people with time for school and work and contributes to improved health for women, children, and families. This machine learning model can help the Tanzanian Ministry of Water improve maintenance operations and ensure that clean, potable water is available to communities across Tanzania. The data for this project was sourced from DrivenData.org from the Pump it Up: Data Mining the Water Table competition (drivendata.org). The data is being supplied to DrivenData.org by the Taarifa waterpoints dashboard, which aggregates data from the Tanzania Ministry of Water. Questions we hope to answer with the data: + Are there any geographic regions with higher pump failure rates? + Which features are most important in determining a pump needs repair? + Are pumps funded by certain funding organizations, installers, or management groups more likely to fail? #### Technologies Used/Plan to Use ## Data Cleaning and Analysis Python is used for initial data cleaning and analysis. Jupyter notebook is used to compose and run the codes. R is used in the further analysis part. ## Database Storage We have used SQLite Database for data storage. Tableau Dashboard along with Google slide is used to display and presentation purposes. ## Machine Learning Python and Jupyter notebook are used for ETL and machine learning. The act …