In this project we made a machine learning model using classic machine learning technics such as random forests and R programming laدguage. This was a project for regression analysis course. The questions answered in this project : 1- Does this data set have any chance for predicting diabetes? 2- Which features have the most power in predicting diabetes? 3- Could we use a subset of the features to make a predicting model ? 4- What is your strategy in using test, train and validation sets to predict the final model? Which features was used in the final model? 5- Could we ask a few questions and predict the risk of diabetes in a person with an acceptable accuracy ?