Title: Predicting housing sale prices in Germany by applying machine learning models and methods of data exploration
Authors: Chong Dae Kim - TH Koeln (Technische Hochschule Koeln) (Germany) [presenting]
Nils Bedorf - TH Koeln (Technische Hochschule Koeln) (Germany)
Abstract: The prediction of real estate prices is a popular problem in the field of machine learning and is often demonstrated. In contrast to other approaches, which regularly focus on the US market, the focus is on the biggest, German real estate dataset, with more than 1.5 million unique samples and more than 20 features. We implement and compare different machine learning models with respect to performance and interpretability to give insight into the most important properties which contribute to the sale price. The experiments suggest that the prediction of sale prices in a real-world scenario is achievable yet limited by the quality of data rather than quantity. The results show promising prediction scores but are also heavily dependent on the location, which leaves room for further evaluation.