Location, location, location: Satellite image-based real-estate appraisal
arXiv:2006.11406
Abstract
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.
References in corpus (5)
- Methods for Interpreting and Understanding Deep Neural Networks
- Model-Agnostic Interpretability of Machine Learning
- Vision-based Real Estate Price Estimation
- Diabetic Retinopathy Detection via Deep Convolutional Networks for Discriminative Localization and Visual Explanation
- AI Blue Book: Vehicle Price Prediction using Visual Features