UCLA

Text mining is a powerful tool that can be used to uncover data that businesses can turn into actionable insights. The rise in technology and real estates online presence has made housing listing data readily accessible. The purpose of this paper is to turn textual data gathered from real estate listings into numerical attributes and fit regression models, alongside other common housing attributes, to predict housing prices. Text mining, machine learning, and statistical techniques were used to transform the dataset, build regression models, and select the best performing model.