The objective is to explore what kinds of employees will have more possibilities to earn more than 50K USD per year using the US Census dataset which contains informations for approximately 300,000 people For each one, we have 42 socioeconomic variables. The steps of cleaning and exploring the data are described in the Cleaning and exploring Data.R file. The steps of The Classification are described in the Classification.R file. I chose to test the classification methods (Decision Tree, Random Forest and Logistic Regression) based on various factors such as Work-class, Age, Sex, Education ,Race and Working weeks.