Location, location, location: Satellite image-based real-estate appraisal
This work addresses the problem of accurate home price estimation for buyers and appraisers, but it is incremental as it builds on existing methods by adding image data.
The paper tackled real-estate price prediction by incorporating satellite images with structured data using convolutional neural networks, resulting in a 7% improvement in MAE over a baseline neural network trained only on structured data.
Buying a home is one of the most important buying decisions people have to make in their life. The latest research on real-estate appraisal focuses on incorporating image data in addition to structured data into the modeling process. This research measures the prediction performance of satellite images and structured data by using convolutional neural networks. The resulting CNN model trained performs 7% better in MAE than the advanced baseline of a neural network trained on structured data. Moreover, sliding-window heatmap provides visual interpretability of satellite images, revealing that neighborhood structures are essential in the price estimation.