Predicting Diabetes with Medical and Demographic Data

Diabetes is one of the leading causes of death worldwide and in the U.S. It affects people of all ages, sexes, ethnicities and socioeconomic levels. However, people who have diabetes might be asymptomatic during the early stage, which means the patient does not feel anything at all. Asymptomatic patients would not get diagnosed until they reach a more serious stage and may miss the best time for treatment. The aim of this project is to collect data on diabetes, analyze the data and use them to build a predictive machine learning model for early-stage diabetes detection. Multiple different data preprocessing and classification methods have been applied and compared for the best prediction accuracy.