Project Overview

In this project, we will apply basic machine learning concepts on data collected for housing prices in the Boston, Massachusetts area to predict the selling price of a new home. We will first explore the data to obtain important features and descriptive statistics about the dataset. Next, we will properly split the data into testing and training subsets, and determine a suitable performance metric for this problem. We will then analyze performance graphs for a learning algorithm with varying parameters and training set sizes. This will enable us to pick the optimal model that best generalizes for unseen data. Finally, we will test this optimal model on a new sample and compare the predicted selling price to your statistics.