The midterm project for the course "R programming for Data Scientists" involved applying various Machine Learning Regression models to the "Ames Housing". The dataset contains 82 attributes and 2930 entries in the dataset. This whole project is done using R-programming language using the R-Studio IDE.
The goal of this application is to perform some data cleaning like removing optional attributes, outlier entries, missing values, then predict the final sale price of houses based on the various features present in the house provided in the form of attributes. The following regression models are used in this application: Multiple Linear Regression, Ridge and Lasso Regression along with Forward and Backward Subset Selection. K-fold cross validation technique is performed on the models to make sure that the models are not over-fitted. After making sure that the models are not over-fitted, we declared that the Lasso Regression is the best model based on the square-root of the Mean Square Error.
Technologies: R-programming, Machine Learning Algorithms
Date: March 2017