House price estimation from visual and textual features

TL;DR

This paper introduces a neural network-based approach that combines visual and textual features from house data to improve price estimation accuracy, outperforming models that rely solely on textual information.

Contribution

The paper presents the first dataset combining house images and textual attributes and demonstrates that integrating visual features significantly enhances price prediction accuracy.

Findings

Adding visual features tripled the R-value.

Decreased Mean Square Error by an order of magnitude.

Outperformed existing textual-only models on benchmark datasets.

Abstract

Most existing automatic house price estimation systems rely only on some textual data like its neighborhood area and the number of rooms. The final price is estimated by a human agent who visits the house and assesses it visually. In this paper, we propose extracting visual features from house photographs and combining them with the house's textual information. The combined features are fed to a fully connected multilayer Neural Network (NN) that estimates the house price as its single output. To train and evaluate our network, we have collected the first houses dataset (to our knowledge) that combines both images and textual attributes. The dataset is composed of 535 sample houses from the state of California, USA. Our experiments showed that adding the visual features increased the R-value by a factor of 3 and decreased the Mean Square Error (MSE) by one order of magnitude compared with textual-only features. Additionally, when trained on the benchmark textual-only features housing dataset, our proposed NN still outperformed the existing model published results.