This document discusses predicting house sales prices based on various factors using a data set containing information on homes in Iowa. It describes performing data wrangling on the data set, which has 79 explanatory variables and converting categorical data to numeric. A correlation matrix is produced showing the correlation between predictor variables and sales price. The results show the model can predict 80% of house sales prices using variables like lot size, quality ratings, year built, basement information, floors, and other features. While the predictions are close, the author notes the model can still be improved to predict prices more accurately and requests comments on how to do so.