A Machine Learning Approach to Delineating Neighborhoods from Geocoded Appraisal Data

Abstract

Identification of neighborhoods is an important, financially-driven topic in real estate. It is known that the real estate industry uses ZIP (postal) codes and Census tracts as a source of land demarcation to categorize properties with respect to their price. These demarcated boundaries are static and are inflexible to the shift in the real estate market and fail to represent its dynamics, such as in the case of an up-and-coming residential project. Delineated neighborhoods are also used in socioeconomic and demographic analyses where statistics are computed at a neighborhood level. Current practices of delineating neighborhoods have mostly ignored the information that can be extracted from property appraisals. This paper demonstrates the potential of using only the distance between subjects and their comparable properties, identified in an appraisal, to delineate neighborhoods that are composed of properties with similar prices and features. Using spatial filters, we first identify regions with the most appraisal activity, and through the application of a spatial clustering algorithm, generate neighborhoods composed of properties sharing similar characteristics. Through an application of bootstrapped linear regression, we find that delineating neighborhoods using geolocation of subjects and comparable properties explains more variation in a property’s features, such as valuation, square footage, and price per square foot, than ZIP codes or Census tracts. We also discuss the ability of the neighborhoods to grow and shrink over the years, due to shifts in each housing submarket.

Influence of the Cadastral Value of the Urban Land and Neighborhood Characteristics on the Mean House Mortgage Appraisal

“…This circumstance can explain why some variables are not significant and why some others are unexpectedly positive or negative. This is corroborated by Ali [62] when stating that the delineation of districts ignores information about the property values found in them.…”

Beyond Proximity: Constructing Organic Neighborhoods Using a Two-Stage Unsupervised Learning Approach

“…Recent methodological innovations suggest opportunities to break the dichotomy between ag-gregation and functional approaches and to exploit data-driven approaches combining elements of both approaches. These approaches use machine learning techniques that are well suited to spatial data and identify similar locations at increasingly smaller geographic scales (Ali et al, 2020). Based on the relatively broad level of census tracts in New York City, Spielman and Thill (2008) use a neural network technique called self-organizing maps to highlight that social and geographic space are correlated.…”

“…Based on the relatively broad level of census tracts in New York City, Spielman and Thill (2008) use a neural network technique called self-organizing maps to highlight that social and geographic space are correlated. Using geocoded point data from three American cities on housing appraisals, Ali et al (2020) use a hierarchical density-based clustering algorithm to delineate local properties clusters each exhibiting similar prices and characteristics.…”

“…To estimate neighborhood boundaries based on physical distance and density, I use the hierarchical density-based clustering algorithm with an application to noise (HDBSCAN) (McInnes et al, 2017;Rahman et al, 2016). Essentially being a density-based clustering algorithm, HDBSCAN is well suited for spatial data like other, purely density-based algorithms (e.g., DBSCAN) (Ali et al, 2020;Rahman et al, 2016). Like DBSCAN, HDBSCAN defines a cluster as a set of points that are within a given radius ϵ from at least one other point in the cluster.…”