Valuing Curb Appeal

Johnson, Erik; Tidwell, Alan; Villupuram, Sriram V.

doi:10.1007/s11146-019-09713-z

Cited by 26 publications

(24 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on our metrical results stated in Table 2, we conclude that adding additional information from the 2 Figure available via: bit.ly/visual_XAI exterior image of a real estate improves the appraisal performance, which supports previous research [6,9,26,29]. However, we find varying performance between the used architectures.…”

Section: Metrical Evaluationsupporting

confidence: 86%

“…Other variables corresponding to vegetation are trees on the lot and within the street, both having a positive price influence [27] because they can reduce traffic noise and spend shadow in the summer [20]. Also, the lawn quality [28] or trimmed vegetation on the parcel [29] are factors that should be considered. They indicate that the house owner has taken care of a high maintenance yard.…”

Section: Real Estate Appraisalmentioning

confidence: 99%

“…They indicate that the house owner has taken care of a high maintenance yard. The absence of a yard and many concrete areas combined with broken sideways are signs for a lower-priced home [29]. One possible explanation could be that these visual signs are often present in unpleasing and less prestigious environments [20].…”

Section: Real Estate Appraisalmentioning

confidence: 99%

“…All these soft factors can be extracted from image data and enhance the feature space about real estate. As previously omitted in the pricing process, they can now increase the predictive performance [29].…”

Section: Real Estate Appraisalmentioning

confidence: 99%

“…As the geographical coordinates are required for the download, we first transform the address of the houses to latitude and longitude with Bing Maps. In a second step, we calculate the bearing from the street to the house, similar to the approach from Johnson et al [29]. We use the direct neighbors on the opposite side of the street as a reference.…”

Section: Data Preparationmentioning

confidence: 99%

See 4 more Smart Citations

Visual Interpretability of Image-based Real Estate Appraisal

Kucklick¹

2022

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

Explainability for machine learning gets more and more important in high-stakes decisions like real estate appraisal. While traditional hedonic house pricing models are fed with hard information based on housing attributes, recently also soft information has been incorporated to increase the predictive performance. This soft information can be extracted from image data by complex models like Convolutional Neural Networks (CNNs). However, these are intransparent which excludes their use for high-stakes financial decisions. To overcome this limitation, we examine if a two-stage modeling approach can provide explainability. We combine visual interpretability by Regression Activation Maps (RAM) for the CNN and a linear regression for the overall prediction. Our experiments are based on 62.000 family homes in Philadelphia and the results indicate that the CNN learns aspects related to vegetation and quality aspects of the house from exterior images, improving the predictive accuracy of real estate appraisal by up to 5.4%.

show abstract