Annals of Operations Research (2025) 344:287–311

https://doi.org/10.1007/s10479-024-06273-1

ORIGINAL RESEARCH

Prototype-based learning for real estate valuation: a machine

learning model that explains prices

Jose A. Rodriguez-Serrano1

Received: 22 December 2023 / Accepted: 3 September 2024 / Published online: 23 September 2024
© The Author(s) 2024

Abstract
The systematic prediction of real estate prices is a foundational block in the operations of
many firms and has individual, societal and policy implications. In the past, a vast amount
of works have used common statistical models such as ordinary least squares or machine
learning approaches. While these approaches yield good predictive accuracy, most models
work very differently from the human intuition in understanding real estate prices. Usually,
humans apply a criterion known as “direct comparison”, whereby the property to be valued
is explicitly compared with similar properties. This trait is frequently ignored when applying
machine learning to real estate valuation. In this article, we propose a model based on a
methodology called prototype-based learning, that to our knowledge has never been applied
to real estate valuation. The model has four crucial characteristics: (a) it is able to capture
non-linear relations between price and the input variables, (b) it is a parametric model able
to optimize any loss function of interest, (c) it has some degree of explainability, and, more
importantly, (d) it encodes the notion of direct comparison. None of the past approaches for
real estate prediction comply with these four characteristics simultaneously. The experimental
validation indicates that, in terms of predictive accuracy, the proposed model is better or on
par to other machine learning based approaches. An interesting advantage of this method is
the ability to summarize a dataset of real estate prices into a few “prototypes”, a set of the
most representative properties.

Keywords Real estate valuation · Non-linear regression · Machine learning ·

Prototype-based models

1 Introduction

Real estate valuation (Pagourtzi et al., 2003; d’Amato and Kauko, 2017) refers to estimating
the price of a property with accuracy on the basis of the current market, and impacts a
wide range of stakeholders across industry and society, such as individuals and organizations
involved in real estate transactions, or policymakers seeking to manage urban development
and housing policies effectively.

B Jose A. Rodriguez-Serrano
[email protected]
1 Department of Operations, Innovation and Data Sciences, Universitat Ramon Llull Esade, Barcelona,
Spain

123
288 Annals of Operations Research (2025) 344:287–311

From an operations research standpoint, the ability of firms to predict real estate prices
with precision is crucial for revenue management (Tchuente and Nyawa, 2022), portfolio
optimization (Amédée-Manesme and Barthélémy, 2018; Kok et al., 2017), or risk manage-
ment (Doumpos et al., 2021) and serves as a foundational step which impacts the operations
of specific sectors, such as insurance or market research.
The progress in digitization and increased data accessibility has enabled the systematic
and large-scale estimation of property prices, known as mass valuation (McCluskey et al.,
2013; Kok et al., 2017), often involving the use of statistical or machine learning techniques.
According to Alexandridis et al. (2019), “automatic mass valuation can reduce operational
costs since it is inexpensive and can be performed in a regular basis”.
A common approach to mass valuation of real estate properties is hedonic pric-
ing (Malpezzi, 2003). Hedonic models decompose the utility of a good in the sum of utilities of
its characteristics independently. An iconic example of a hedonic price model is OLS regres-
sion (James et al., 2013), wherein the unit value assigned to each characteristic corresponds
to the regression coefficient. OLS has been widely applied to real estate valuation (Ottens-
mann et al., 2008) and possesses numerous benefits. First, it can be considered a supervised
learning model: it can be adjusted with training data to optimize any given loss function,
and later be used to predict the price of new properties. Second, it is easily interpretable:
the price directly decomposes as a linear combination of the input variables. Third, OLS is
a well-established, accessible and easily reproducible methodology.
Nevertheless, the OLS methodology for price prediction is not without common criticisms.
These include its limited ability to handle intricate, non-linear associations or the risk for
misleading coefficient interpretations when OLS assumptions are violated. Another drawback
of OLS, which is crucial for real estate modeling, is that incorporating spatial coordinates is
not straightforward. While technically feasible, it is unlikely that variables such as longitude
and latitude have linear relationship with price. This is the reason why some hedonic models
tend to incorporate manually constructed variables describing location (Tchuente and Nyawa,
2022) or other factors related to location, such as quality of the views (Potrawa and Tetereva,
2022) or employment potential (Ottensmann et al., 2008).
One step further in sophistication, state-of-the-art machine learning (ML) techniques,
such as random forests (Breiman, 2001) or neural networks (Goodfellow et al., 2016) have
attracted the attention of researchers in real estate valuation (Doumpos et al., 2021; Peter-
son and Flanagan, 2009). Their advantages include their capability to deal with non-linear
relationships, which normally translates to superior performance. Importantly for real estate
research, some models like decision trees, random forests or neural networks can accept
location variables such as longitude and latitude without any further processing (Tchuente
and Nyawa, 2022).
However, there is a prevalent consensus that while these systems excel in prediction
capabilities, they often fall short in providing valuable insights due to their inherent lack of
interpretability. Notably, the main criticism is the difficulty to assign a meaning to the model
parameters. To counterbalance these limitations, some works in real estate valuation have suc-
cessfully combined machine learning models with explainable AI (XAI) techniques (Lorenz
et al., 2022; Potrawa and Tetereva, 2022). These techniques complement machine learning
models by extracting metrics such as relevance scores for each variable. Still, an issue is that
these techniques are applied ex post and remain detached from the model itself.
Furthermore, one additional limitation of both OLS and ML models is that they ignore
a fundamental cue in valuation theory: direct comparison—perhaps the most common and
intuitive way in which humans appraise real estate. In direct comparison, “the value of
the property being appraised [...] is assumed to relate closely to the the selling prices of

123
Annals of Operations Research (2025) 344:287–311 289

selling properties within the same market area”, as put by Pagourtzi et al. (2003), who name
this concept the comparables method. The same idea has also been dubbed direct capital
comparison (DCC) by Lins and L.F.d.L. Novaes, and L.F.L. Legey. (2005). In order to unify
the terminology, in the following we refer to this concept as direct comparison.
If the models used for real estate valuation incorporated the notion of direct comparison,
they would be better aligned with human expectations and market domain knowledge. Yet
it can be clearly argued that OLS and, more broadly, machine learning approaches, do not
implement explicitly the idea of direct comparison, because these models establish direct
mappings from input variables to prices. In other words, most of these predictive models
do not involve similarity computations between the input property and the nearby or similar
properties.
There are notable exceptions to the previous statement, such as K-Nearest Neighbors
(KNN) (Kramer and Kramer, 2013) and its variants. These are designed to offer predictions
grounded in such comparative analyses and have exhibited good performance in the domain
of real estate price prediction (Choy and Ho, 2023; Isakson, 1988). One technical drawback
of KNN is that it is a non-parametric density estimation method; as such it cannot be adjusted
to minimize a loss metric. The only mechanism to increase the quality of the estimation is
by increasing K , which renders the method inefficient.
Based on this analysis, it is clear that none of the discussed approaches simultaneously
complies with one of these characteristics: (i) capability to model complex and non-linear
relationships, (ii) capability to optimize a predictive objective function, (iii) exhibit some
degree of explainability, and (iv) implement the notion of direct comparison explicitly.
In this article, we propose a method that complies with these four characteristics simulta-
neously. The method is based on a methodology known as “prototype-based learning” (Biehl
et al., 2016) which, to the author’s knowledge, has not been previously applied for real estate
valuation. Prototype-based learning is a predictive model that:
• can learn non-linear relations between the dependent and input variables,
• is a supervised learning method that can be trained to minimize any given loss function
(unlike KNN),
• includes explainability as part of the model, and
• implements explicitly the notion of direct comparison: the model predicts the price of
a property by contrasting the property of interest to a set of (automatically inferred)
reference properties.
Table 1 summarizes these four characteristics and compares the aforementioned models
and approaches in terms of which of them they fulfill.
The model works by minimizing a prediction error, based on comparing the variables
of a property with the distributions of a set of references, and then optimizing these refer-
ences. Therefore, as a by-product, the model produces a segmentation of the data into groups
(or areas) with differentiated characteristics, similar to sub-market models (Goodman and
Thibodeau, 2007; Rey-Blanco et al., 2024).
We study prototype-based models using a public dataset of real estate prices in Seattle. Our
experiments conclude that a) prototype-based learning successfully distills the prototypes as
theoretically expected, b) the method outperforms other strategies for finding prototypes, c)
the prediction error observed is lower compared to other machine learning models, and d)
the model can incorporate location-based information in an effective way.
The rest of the article describes the model (Sect. 3) in detail, reports the different exper-
iments (Sect. 4), provides a detailed discussion (Sect. 5), reviews the practical and research
implications (Sect. 6) and comments on conclusions and future perspectives (Sect. 7).

123
290 Annals of Operations Research (2025) 344:287–311

Table 1 The columns show the design desiderata for our real estate prediction model as discussed in the main
text
Non-linear Parametric training Explainable Implements DCC

OLS  
ML  
ML + XAI   
KNN   
Prototype-based    
The rows specify which criteria are met by each family of approaches: ordinary least squares (OLS), state-
of-the-art machine learning models (ML), ML with explainable AI (XAI) techniques, K-Nearest Neighbors
(KNN) and prototype-based model, the method under study in this article. The latter is the only approach that
fulfills all the desiderata

2 Related works

2.1 Hedonic models and OLS

The origins of the hedonic pricing theory have been a subject of debate between schol-
ars (Goodman, 1998; Colwell and Dilmore, 1999). It is often acknowledged that the first
use of “hedonic” is by Court (1939). However, Haas (1922) already proposed to predict land
price per acre based on a linear model more than one century ago.
As of today, OLS is still a common method to model, and predict, land or real estate prices.
We refer the readers to Hong et al. (2020) and Potrawa and Tetereva (2022) for comprehensive
reviews of prior research that adhere to this methodological approach and delve into various
categories of employed variables. A well-established consensus, substantiated by empirical
evidence, underscores the positive correlation between property prices and factors such as
floor area, the number of bedrooms, and bathrooms. In contrast, certain variables, such
as property age, exhibit a negative association with property prices (Hong et al., 2020).
It is important to consider factors other from those describing the physical property, such
as “locational” or “neighborhood”-based factors (Potrawa and Tetereva, 2022) which tend
to exhibit a strong influence, too. For instance, Ottensmann et al. (2008) and Hussain et
al. (2019) study the impact of urban factors, such as proximity to employment locations,
or impoverished neighborhoods, respectively, in real estate prices. Some works even go
as far as using variables encode the quality of the view, as extracted by image analysis
techniques (Potrawa and Tetereva, 2022) or quality scores extracted from user reviews (Gibbs
et al., 2018).
OLS finds advantages in its simplicity and ability to interpret the coefficients as the
average increase in price after a one-unit increase of the corresponding variable. However,
some drawbacks include the limitation of the functional form, incapable to model non-
linear relations between variables, interactions or colinearities (Zurada et al., 2011). Another
common drawback is the usual assumption of linear distribution, rarely observed in prac-
tice (Potrawa and Tetereva, 2022). Despite these (well-known) technical limitations, OLS
still finds many applications, as some practitioners find practical productivity increases in
predictive applications.

123
Annals of Operations Research (2025) 344:287–311 291

2.2 ML models

As an alternative to OLS, previous studies have resorted to more advanced machine learning
techniques in order to obtain models with more complex functional form.
This includes the use of neural networks (Peterson and Flanagan, 2009; McCluskey et
al., 2013), which produce non-linear predictions by building a chain of linear operations and
non-linear activation functions, and are considered universal approximators (Goodfellow et
al., 2016). Other common approaches to produce robust non-linear characteristics are the
random forests (Hong et al., 2020) or other forms of tree-based ensembles (McCluskey and
W., D. Zulkarnain Daud, and N. Kamarudin., 2014; Kok et al., 2017). These models are
based on regression trees, which produce predictions based on a sequence of variable tests,
capturing non-linear relations and interactions between variables. Then they average many
regression trees using different criteria to increase robustness.
Neural networks and tree-based ensembles are probably the two most common state-of-
the-practice approaches in machine learning. We refer the reader to Doumpos et al. (2021)
or Tchuente and Nyawa (2022) for an exhaustive review and comparison between different
machine learning methods in real estate modeling.
Because of their added complexity, machine learning models tend to be, on the one hand,
more accurate than OLS. On the other hand, an oft-cited issue of most ML approaches is
their lack of interpretability. Interpretability denotes the capability to easily understand and
describe how an output is produced given an input, and is an important feature for market
regulators, or in general to augment the trust in decision-support systems. While some ML
models are considered white-box, such as decision trees, most state-of-the-art ML models,
such as those based on neural networks or tree ensembles are considered black-box (Tchuente
et al., 2024). To overcome that limitation, there has been a recent emergence of explainable AI
(XAI) techniques (Lundberg et al., 2020). In real estate prediction, Lorenz et al. (2022) have
applied XAI methods such as partial dependence plots and feature importance to complement
ML-based models, the latter also being used by Potrawa and Tetereva (2022). XAI methods
are employed to diagnose black-box models, often are applied ex post, and are not part of
the model. In contrast, as acknowledged by Li et al. (2018) (and further explained in Sect.
3), prototype-based models provide explanations naturally as part of the model.

2.3 Towards direct comparison: KNN and clustering

Among specific machine learning models used in real estate research, those that take into
account direct comparisons by design include those based on KNN (Choy and Ho, 2023;
Isakson, 1988; Tchuente and Nyawa, 2022). In KNN, the regression is taken as a weighted
average of the outputs of the nearest data points of the training set, based on some distance
(usually, Euclidean) between the vectors of input variables.
As such, KNN is capable of modeling complex, non-linear effects between price and the
input variables. Its advantages also includes its methodological simplicity and interpretability.
However, one drawback of standard KNN on the predictive performance side is that it is a
non-parametric model and cannot be adjusted based on a target loss function.
In order to combine the benefits of KNN and other machine learning methods, some
real estate studies perform spatial weighting schemes (Bitter et al., 2007; McCluskey et al.,
2013), of which a well-known type is geographically weighted regression (Fotheringham et
al., 2003). Spatial weighting assigns more importance during model training to those samples
which are proximal to the example whose price is to be predicted.

123
292 Annals of Operations Research (2025) 344:287–311

The model proposed in this paper addresses both characteristics: first, it uses comparative-
based measures, akin to KNN, while offering the flexibility to be fine-tuned based on a
user-defined loss function. Experiments demonstrate that this crucially reduces the prediction
error. Second, it incorporates spatial weighting as an intrinsic feature.

2.4 Prototype-based learning

In summary, the advantages of prototype-based models in terms of modeling non-linear rela-

tions, explainability, availability of comparison-based valuation, and ability to be optimized
given a loss function make those models very promising for predicting real estate prices.
The term prototype-based models (Biehl et al., 2016) encompasses a family of approaches
rather than a singular methodology, and within this family we focus on the subset of super-
vised learning approaches. These make predictions by leveraging latent “prototypes”, which
refer to a reduced set of data samples which summarize most of the patterns occurring the
whole dataset. Prototypes can correspond to actual data points, or can consist of calculated
variables and not strictly correspond to any real observed datum. Additionally, when the
space is segmented based on proximity to prototypes, these models facilitate the identifica-
tion of distinct regions or clusters within the data space, effectively acting as “cluster-aware
supervised learning algorithms” (Chen and Xie, 2022).
The next section is devoted to describing the specific method employed in the paper
and will introduce many technical aspects of prototype-based methods. This methodology
finds applications in diverse areas such as computer vision (Li et al., 2018), natural sci-
ences (Owomugisha et al., 2021), or healthcare (van Veen et al., 2022), including COVID
detection (Kaden et al., 2022). To the best of the author’s knowledge, however, this method
has never been applied to real estate prediction, in spite of its appealing benefits.

3 Model

3.1 Notation

The proposed model requires a dataset of real estate properties with N samples, each with
D independent or (in the following) input variables xnd (thus n ∈ {1, . . . , N } and D ∈
{1, . . . , D}) describing characteristics of the real estate properties, and a dependent or output
variable yn indicating the observed price. By price we mean either the direct price, or any
transformation of the price, such as the price normalized by a constant or non-constant factor
(e.g. price per square foot), or the logarithm of the price, just to mention some commonly
used options. Our model is compatible with all these definitions.
We will use the boldface form xn , common in the ML literature, when we have to refer
jointly to the set of all input variables xn1 , . . . , xn D of sample n. We will omit the index n
when it is not necessary for the discussion. In a slight abuse of language, occasionally the text
may refer to x shortly as a “property” (instead of “the values of the input variables describing
a property”).

3.2 Basic model structure

The main design principle of our model is that it should implement the notion of direct
comparison: if we observe a new property with input variables x, we should be able to infer

123
Annals of Operations Research (2025) 344:287–311 293

its price by comparing it to a number of “references”, denoted as prototypes, and interpolating

from the price of similar references.
Next, we discuss how the model proposed in this article directly implements this principle.
We define a set of M prototypes, which represent a set of properties with similar input
variables and with an attributed price. Formally, each reference consists of:
1. A prototype distribution, which models the probability distribution f m (x) of properties
within group m. In the simplest case, we take a multivariate Normal distribution
f m (x) = N (x|pm , σ m ) (1)
where pm = ( pm1 , . . . , pm D ) specifies a vector of means and σ m refers to the vector of
standard deviations (assuming diagonal covariance).1 Because pm represents the mode
of f m , it therefore indicates the typical values of the input variables for each prototype.
2. A prototype value vm , which models an price attributed to that group of properties.
To compare a property x to each of the M references, a principled method (Bishop, 2006) is
determining the posterior probability that f m has generated x, i.e.
f m (x)
γm (x) ≡  . (2)
l fl (x)
This comparison should yield, ideally, a small set of properties with non-negligible values of
γm (x) which represent the most similar references. Then the predicted value is a a weighted
average2 of the references, where the weighing coefficient is precisely the value of γ :

M
ŷ = γm (x)vm . (3)
m=1
As we will verify empirically later, it is common that most values of γm (x) are negligible
and only a few terms contribute to the sum, and therefore Eq. 3 can be understood as an
“interpolation” of the attributed values of the most similar references.
Note that the proposed model is completely determined by pmd and vm (and also σmd in
case these are optimized). We jointly denote these parameters as θ . Its optimal values will be
learned from the data using the following loss function:

1 
N
C= (yn − ŷn )2 + η R(θ ), (4)
N
n=1
The first term is as mean squared error (MSE), which measures the fitness error of the model
with respect to the data, and R is a regularization term, common in ML tasks to penalize
model complexity or to impose reasonable restrictions. In our case, we use
1 
R(θ ) = min pm − xn . (5)
M m n
This term averages the distance between each prototype and its closest data point, similar
to Li et al. (2018), therefore adding a penalty if prototypes pm are far from real properties
x(here,  ·  stands for the Euclidean norm).
1 A simplification used in some cases is to assume the same value of the standard deviation for all variables
(isotropic covariance) and all prototypes, in which we will refer to the standard deviations as the scalar σ .
Moreover, in that simplified case, σ can be fixed a priori and treated as hyperparameter (which sometimes
improves the numerical stability of the model).
2 Because, by definition,  γ (x) = 1, Eq. 3 can be interpreted as a weighted average.
m m

123
294 Annals of Operations Research (2025) 344:287–311

Fig. 1 (Best viewed in color) Illustrative example of the prototype-based method. (Color figure online)

It is straightforward to identify that Eq. 4 is differentiable with respect to θ . Therefore,

we can use standard solvers that involve automatic differentiation (Baydin et al., 2018) such
as keras to implement and this model and find its optimal parameters.

3.3 Model interpretation and simple example

The parameters pmd , vm of the model have a natural interpretation. Specifically, as pmd is the
mean of the distribution f m , it can be interpreted as the variables of a representative property,
and vm as its representative price. Therefore, not only will the model be able to predict
property prices, it also produces a list of M prototypical properties and suggests values for
their expected variables and price.
To illustrate the model capabilities, for now we select a simple example on a controlled
setting (experiments on real data are reported in the next section). Figure 1 shows a scatterplot
corresponding to a small dataset with two input variables. The locations of the circles on the
horizontal and vertical axes represent, respectively, the input variables x n1 and xn2 , and the
color indicates the value of the output variable y (yellow for y = 0.0 and blue for y = 1.0).
In this case, we train the prototype-based model with M = 20, and using a loss function
with η = 0. In this setting, the parameters can be directly translated to the plot: pm1 and pm2
represent the location of the prototype means on the horizontal and vertical axes. The color
indicates the prototype values vm (black for vm = 0 and white for vm = 1). The color scale
is continuous, although perhaps that is not perceivable as most prototype values are close to
vm = 0 or vm = 1.
By visual inspection, it is clear in this example that the locations pmd of the prototypes
span the different “clusters” in the data and the values vm match the values of yn in each area.
As a conclusion, we have quantitative evidence in a controlled example that the obtained
prototypes can be interpreted as representative property characteristics and their price.
To shed light on how the model implements the intuition of direct comparison, consider
the prediction for a given sample x. To apply Eq. 3, first we need to determine the values
of γ in Eq. 2. Continuing with our example, in the case of two variables, using a bivariate
Normal distribution for f m , we have:

1 (x1 − pm1 )2 + (x2 − pm2 )2

log f m (x) = − log(2π) − 2 log σ − . (6)
2 σ2

123
Annals of Operations Research (2025) 344:287–311 295

Fig. 2 a Prototype-based model interpreted as a neural network. b Example of extending the prototype-based
model neural network with an additional hidden (embedding) layer

If we define λm ≡ log f m (x), then Eq. 2 becomes

eλm
γm (x) =  λ , (7)
ke
k

sometimes known as the softmax function.

As we can verify from Eq. 6, as the Euclidean distance from (x1 , x2 ) to ( pm1 , pm2 )
decreases, the value of γm increases and the contribution of the value vm becomes higher.
Crucially, this is precisely the expected behavior in a comparison-based model.

3.4 Model extension as neural network

The prototype distribution f m in Eq. 1 takes as input a property x. However, we note that
our model is compatible with approaches that first compute a so-called embedding h =
g(x), where g denotes the result of neural network layers applying some transformation to
the input variables. In this section, we demonstrate how our model can be interpreted as a
neural network and integrated into standard neural network frameworks. We utilize common
neural network terminology throughout. Readers who are not familiar with neural network
models can skip this section without impacting their understanding of the rest of the article.
Alternatively, they may refer to foundational texts such as Goodfellow et al. (2016) for further
information.
First, note the sequential computation of Eqs. 2 and 3 can be expressed in neural network
vocabulary:

• Eq. 2 is equivalent to passing the inputs x through a probabilistic custom layer with M
units, where the mth unit outputs log f m (x), and then applying a standard “softmax”
activation function.
• Eq. 3 is simply a dense layer, which takes as input the previous outputted γ ’s, multiplies
them with some weights and outputs ŷ. Since there is only 1 output, there are m weights
which are precisely the prototype values vm .

A schematic representation of the prototype-based model interpreted as a neural network

is depicted in Fig. 2a.
This interpretation as neural network has two main benefits. First, it facilitates the code
implementation of our model, as the previously described network can be implemented

123
296 Annals of Operations Research (2025) 344:287–311

using commonly available libraries such as tensorflow or keras (see Sect. 4.2 for
implementation details).
A second benefit is that we can add further hidden layers between the inputs and the custom
probabilistic layer. This adds a (potentially non-linear) transformation of the input variables
and thus increases the complexity of the predictive function. An example where a hidden
layer (embedding) is added to our model is illustrated in Fig. 2b. With this architecture, now
the input to the prototype-based model is the output of the hidden layer.3
In such a case, the parameters of the neural network g can be pre-trained, or jointly
optimized during the training of the prototype-based model. Technically, one needs to
replace f m (x) by f m (h) and this turns the loss function in Eq. 4 an implicit function of
g and its weights. This change is straightforward to apply in the aforementioned automatic
differentiation tools.
One drawback of this extension is that the prototype-based model becomes non-
interpretable if additional layers are added. However, in some cases, prioritizing predictive
accuracy over interpretability may justify this approach. Nonetheless, this extension is
optional, and we simply emphasize that our model is compatible with it.

4 Experimental validation

4.1 Data

We use data from a record of properties in King County (Seattle),4 which includes property
prices and other variables collected from May 2014 to May 2015.
The dataset contains information about 21,613 properties. Table 2 indicates a summary
of the variables available in the dataset that will be relevant for the analyses below, and some
typical statistics. Fig. 3 displays a map of the location of the properties , where the color
indicates the price per square foot.

4.2 Model implementation

The model is implemented in the Python programming language, using keras as the
automatic differentiation engine. The code is publicly available.5

4.3 Model using location variables only

The first experiment will evaluate the model considering exclusively location and price per
square foot. There are several reasons for this choice. First, it is widely acknowledged that
“location is the primary factor in determining market prices” (Gabrielli et al., 2023) and
that this primacy arises from its multifaceted influence on factors such as proximity to eco-
nomic opportunities, appeal of certain leaving areas or transportation accessibility. Ultimately
“virtually every attribute of a piece of land beyond its physical dimensions including any
3 According to our previous definition h = g(x), in this simple case of 1 hidden layer, g is simply g(x) =
α(Wx), where W denotes a matrix of weights and α an activation function.
4 Publicly available at https://geodacenter.github.io/data-and-lab/KingCounty-HouseSales2015/
5 The code is structured as follows: model and examples are available in https://github.com/
prof-jose/prototype_based_learning and the experimental benchmark in https://github.com/prof-jose/
prototype_based_learning.

123
 Table 2 Summary of the variables used for the case of the King County properties dataset
Variable Data type Statistics
Min Max Mean Std

Date Date 2014/05/02 2015/05/27 N/A N/A

Latitude Numeric 47.2 47.8 47.6 0.139
Longitude Numeric −1.23 · 102 −1.21 · 102 −1.22 · 102 0.141
Number of bathrooms Numeric 0 8 2.11 0.770
Annals of Operations Research (2025) 344:287–311

Square feet of living room Numeric 2.90·102 1.35·104 2.08·104 9.18·102

Square feet of lot Numeric 5.20·102 1.65·106 1.51·104 4.14·104
Number of floors Numeric 1.0 3.5 1.49 0.540
Waterfront Binary 0 1 7.54·10−3 8.65·10−2
View Numeric 0 4 0.234 0.766
Square feet above Numeric 2.90·102 9.41·103 1.79·103 8.28·102
Square feet of basement Numeric 0.0 4.82·103 2.92·102 4.43·102
Square feet of living room in 2015 Numeric 3.99·102 6.21·103 1.99·103 6.85·102
Square feet of lot in 2015 Numeric 6.51·102 8.71·105 1.28·104 2.73·104
Price Numeric 7.50·104 7.70·106 5.40·105 3.67·104

123
297
298 Annals of Operations Research (2025) 344:287–311

Fig. 3 (Best viewed in color) Locations of properties in out dataset on a real map. The color intensity indicates
the quantile of the price per square foot (darker indicates higher). (Color figure online)

improvements can be reduced to location” (Atack and Margo, 1998). Therefore, we simu-
late a model that predicts real estate price solely based on location. Second, when the input
variables are location, the prototypes pm of our model can be directly interpreted as a list
of typical locations, and the values as typical prices in these locations. While deliberately
simplistic, this is still very relevant information for practitioners in real estate or seeking to
obtain a geographical segmentation of the market and sub-market analyses (Goodman and
Thibodeau, 2007).
Experimental protocol
To perform model training and evaluation, an initial step involves a linear partition of
the dataset into training and testing subsets (with a training/test distribution of 60%/40%),
structured in accordance with chronological order (given by the date column). This approach
emulates real-world conditions where available information is time-bound, and the goal is to
predict property values beyond a specific temporal point. This method effectively mitigates
the potential leakage of temporal information into the predictive models. Notably, within the
training subset, 20% of the data is reserved for parameter validation, ensuring that model
parameters are optimized effectively.

123
Annals of Operations Research (2025) 344:287–311 299

Table 3 RMSE of the models

Method RMSE (Test) (ft2 )
using only location variables,
averaged over 10 runs Random prototypes 110
K-Means + Price Average 81.7
Prototypes 76.6

For the analysis, we take latitude and longitude as the input variables, and construct an
output variable price per square foot as the quotient of the available columns price and square
feet.
In the data pre-processing stage, each input variable is standardized (to zero mean and unit
variance). The output variable y (price per square foot) is scaled by a factor of 1/1000, with
the objective of rescaling the values of y to fall within the approximate range of (0.0, 1.0),
which makes numerical optimization more favorable.
To build our model, we use M = 50 prototypes. The model is implemented and optimized
using the keras library, with an Adam optimizer with learning rate lr , batch size b and
number of epochs E. To these hyperparameters, we add the regularization coefficient η of
Eq. 4 and also use a constant preset σ for n Eq. 1. All these hyperparameters are systematically
determined using a standard grid search strategy on the validation set. We highlight that we
use a constant value of σ for both coordinates and for all prototypes, so that the computation
of importances γ of Eq. 2 is explicitly a function of the Euclidean distance between each
property and the prototype mean.
Our model is subjected to comparative evaluation against two alternatives. First, we con-
duct a comparison with a nearest neighbor algorithm using randomly selected neighbors and
M=50 (in order to match the number of prototypes). This comparison offers insights into
the efficacy of the prototypes in guiding the predictive process. Second, we compare with an
alternative way of discovering prototypes, wherein we use a clustering algorithm. Clustering
stands as an alternative for identifying groups of properties with similar characteristics and
valuations. Therefore, we apply a K-Means clustering algorithm (Bishop, 2006) with K = 50
and take its means as the prototypes pm , and the average price per square foot in each cluster
as the vm . For all models, we measure the quality of fit with the root mean squared error
(RMSE), in the original y-scale (before normalization).
To mitigate the impact of sources of variability such as random seeds used in parameter
initialization, we repeat the experiment a total of 10 times, and report the RMSE averaged
over the 10 runs.
Quantitative results
We first evaluate the predictive accuracy of the model. The RMSE values of the different
models tested in experiments described above are shown in Table 3. We observe the prototype-
based model yields significantly lower RSME than the two baseline methods (the standard
deviation of the RMSE over the 10 runs is smaller than 1 $/ft2 ), therefore indicating that it
has accurate prediction capabilities.
Results: Interpretation of the prototypes
Now we turn to discussing the interpretation of the prototypes pm obtained by the model.
Because our input variables are longitude and latitude, the model parameters pm and vm can
be interpreted directly as the prototype locations and prototype values. Figure 4 displays the
locations pm overlaid on the original data.

123
300 Annals of Operations Research (2025) 344:287–311

Fig. 4 (Best viewed in color) Locations of the prototypes pm overlaid in the original data, and showing the
corresponding vm for a specific example. (Color figure online)

Fig. 5 Zoom into one of the prototypes. (Color figure online)

Through our analysis, it becomes evident that the prototypes pm exhibit a wide distribution
that effectively spans the different regions of the entire dataset. Furthermore, as depicted
in Fig. 5, a closer examination is directed towards a specific prototype, shedding light on
properties in close proximity to it or, in other words, all the properties that can be succinctly
encapsulated or “summarized” by the respective prototype.
We verify by visual inspection that the values vm and the nearby prices are consistent.
Therefore, we qualitatively verify that the behavior of the model mimics the process of
valuation-based comparison that is similar to the human intuition behind pricing.
Also, we quantitatively investigate how many prototypes typically contribute to each
prediction when evaluating Eq. 3. To that end, we compute the prototype importances γm (x)
for all properties x of the test set, and count the non-negligible terms as those with γm (x) >
0.01. Table 4 displays the resulting frequencies. Notably, in no case did we observe more

123
Annals of Operations Research (2025) 344:287–311 301

Table 4 Frequencies of the Non-negligible prototypes Frequency (%)

number of non-negligible
prototypes contributing to the 1 74.8
prediction in Eq. 3
2 19.8
3 4.7
4 0.7
5 < 0.05
6 and more 0
All the predictions involve less than 5 non-negligible prototypes, and
more than 94% involve only 1 or 2 prototypes

Fig. 6 Intepretation of the model as a zonal segmentation: each zone represents an area with approximately
constant prices, where that constant is given by the obtained values vm . Two examples are shown: a vm ≈ 424,
b vm ≈ 165

than 5 prototypes with γm (x) > 0.01. Furthermore, for over 94% of the test samples, only 1
or 2 prototypes significantly contributed to the prediction.
Interpretation as zonal segmentation
The model’s output can alternatively be interpreted as a zonal segmentation: defining the
mth “zone” as the area where all properties are closest to the mth prototype results in a
Voronoi segmentation, as illustrated in Fig. 6. Each zone represents a geographical area with
similar property prices, where the reference price is specified by the prototype value vm . In
Fig. 6, two examples of vm are depicted to illustrate this concept.
This interpretation is connected to research in sub-market segmentation. According to
Thibodeau (2003), “a housing [sub-]market defines a geographic area where the price of
housing per unit of housing service [...] is constant”. Therefore, our model could be interpreted
also as a means to perform sub-market segmentation, since each prototype m roughly defines
an area where, in its vicinity, the price is approximately a constant vm .
It is noteworthy that many traditional approaches to sub-market segmentation rely on
grouping predefined market-based or policy-based elemental areas, such as postcodes or
school districts (see, for example, Table 1 in Borst and McCluskey (2008)). In contrast, our
approach infers sub-markets automatically from raw data using a machine learning algorithm.
This rather aligns us with recent works such as Rey-Blanco et al. (2024).

123
302 Annals of Operations Research (2025) 344:287–311

Fig. 7 RMSE of the different

compared methods for different
number of prototypes

Fig. 8 Scatterplots for the 12-variable case, showing pairs of variables and the obtained prototypes

In any case, we emphasize this capability to produce market segments automatically as a

practical by-product of the proposed model. An in-depth analysis of prototype-based models
to the specific task of submarket segmentation would require a separate study, which we
defer it to future work.
Results: On the effect of the number of prototypes
We now study the effect of the number of prototypes. We repeat the previous protocol
using different number of prototypes and compare the three different methods with M =
10, 20, 50, 100. Results are reported graphically in Fig. 7.
As expected, the error decreases for all methods with more prototypes but the proposed
approach gives the smallest errors across all values of M. Also, if we observe how the error
increases by decreasing M, our model degrades with a smaller slope compared to the other
models.

4.4 Model with more variables

Following the protocol established in the last section, we now incorporate a more comprehen-
sive set of variables. Apart from latitude and longitude, we consider number of bathrooms,
square feet of living room, square feet of lot, number of floors, waterfront, view, square feet
above, square feet of basement, square feet of living room in 2015, square feet of lot in 2015.
Thus, the new model uses 12 input variables in total.

123
Annals of Operations Research (2025) 344:287–311 303

Table 5 RMSE of the compared ˆ

Method RMSE Min RMSE Max RMSE
models using 12 input variables
Random prototypes 115 107 125
kmeans + Price Average 90.9 90.2 91.6
Prototypes (constant σ ) 68.1 66.9 69.5
Prototypes (fitted σ ’s) 62.0 60.7 63.2

Fig. 9 Comparing the distribution of test prices with the distribution of estimated prices, for M = 50 (left)and
M = 100 (right) prototypes

The experimental details are the same as in the last section, with one exception: because
now we use several variables of different types, it is reasonable to expect that different
variables need to be weighted differently for the importance computation of Eq. 2. Our
model also accepts this design choice, which translates to fitting the values of σm of Eq. 1. For
completeness, we will report the results using a constant hyperparameter σ for all prototypes
(equivalently to the last section), and also by explicitly learning the σmd ’s as part of the
optimization process.
The results are summarized in Table 5. Again, the table displays the values of RMSE
for the same methods of the previous section. In this case, we found the random neighbor
initialization of the KNN method yields a higher variance between the 10 runs. Therefore,
we also report the minimum and maximum RMSE across the 10 runs to make clear there is
no overlap in the error metrics.
There are some noteworthy observations when comparing to the results of the previous
section, which leveraged location variables only. First, the RMSE of the prototype based
method (using constant σ ) has improved (down to 68.1 in Table 5 vs. 76.6 of Table 3). This
underscores the positive impact of adding more variables in the prototype-based model. This
impact is even larger when one considers the improvement brought by learning the parameters
σm (down to 62.0).
We can still interpret the means pm as the prototype modes—with the clarification that
now these do not only specify the typical locations, but also the typical values of each variable
considered. As an illustration, Fig. 8 displays six scatterplots, covering the 12 input variables,
and also the projections of the prototypes in the corresponding subspace.
Second, when inspecting the results for the other 2 methods, we observe an opposite trend:
in both cases the RMSE has not improved but increased as more variables are integrated (for
K-Means, RMSE increases to 90.9 using the 12 input variables, up from a value of 81.7 with
the model of the previous section; for KNN the increase is from 110 to 115).

123
304 Annals of Operations Research (2025) 344:287–311

Fig. 10 Prototypes (red) obtained for the neural network embeddings of properties (blue). (Color figure online)

A possible explanation for this phenomenon is as follows. When we used latitude and
longitude as the only input variables, the proximity to a prototype (both KNN and K-Means
are based on Euclidean distance) is a good proxy for the similarity in price, as argued previ-
ously. However, when additional variables are introduced, the distance calculation between
a property and the prototype becomes a composite measure, encompassing multiple compo-
nents, some of which may not have a direct correlation with price differences. In essence,
this expanded feature space introduces dimensions that are not necessarily indicative of vari-
ations in property prices. Consequently, similarity between variables does not translate to an
effective similarity in price.6 The reason why our prototype-based model exhibits the oppo-
site behavior can be attributed to the model’s ability to adjust the optimal locations of the
prototypes (and, especially, the scales of each dimension). By doing so, it effectively hones
in on the specific variables and combinations thereof that are most relevant for accurate price
estimation.
Approximating the target distribution
Another research question we could ask is do the model predictions ŷ also approximate
well the distribution of observed test prices y? To that end, Fig. 9 displays the histogram
of both test and estimated prices. For the case of M = 50 (the setting used in this section),
indicated on the left, we see the distribution of estimated prices substantially overlaps that
of real prices, albeit putting more mass in the center of the distribution and less in the
extremes. However, this phenomenon is not surprising and, to some extent, expected, since
prediction models can be interpreted as “smoothing the data” (and consequently reduce

6 For instance, consider the K-Means method and property #4607 from the test set, which shows the largest
error difference between the location-only model and the model with all features. This property has normalized
coordinates (−1.62, −1.05) and a normalized price of 0.118. The nearest prototype in the location-only model
is at location (−1.81, −1.05) with a value of v9 =0.149. In the model with all features, the closest prototype is
at (−0.55, −0.83), but the proximity has been driven by very similar values of the “waterfront”, “floors” and
“bathrooms” variables (and clearly not location). However, this prototype’s normalized value is v5 = 0.44,
which deviates significantly from the actual price. This is an example where similarity in specific variables
does not necessarily imply a price similarity. The K-Means algorithm creates clusters (prototypes) without
considering variable importance or property prices, leading to errors for properties like #4607 in the all-features
model, which we verify is not an isolated case.

123
Annals of Operations Research (2025) 344:287–311 305

Table 6 Comparison of the Method RMSE Num. parameters Explainable

RMSE of the different models
(lower is better) Regression tree 67.6 1748 YES
Random forest 65.8 1740 NO
Neural net 61.9 1321 NO
Prototypes 62.0 1250 YES

extreme observations). To verify this, if we increase M, as we use more prototypes we

should expect less smoothing, which is clearly visible in the plot on the right for M = 100.

4.5 Comparison with other predictive models

So far, the prototype-based model has shown some interesting advantages in terms of predic-
tive power and also in terms of explainability. And the previous sections have demonstrated
that it outperforms other alternative ways of obtaining prototypes, both in the location-only
case and when using multiple variables.
Now we compare the prototype-based model with three models discussed in Sect. 2: (i)
regression trees, (ii) random forests and (iii) a standard neural network based on a multi-layer
perceptron.7
First, we discuss the comparison settings. For all methods, we still use identical
train/validation/test splits as in the case of the prototype-based model. We select the best
hyperparameters by evaluating the RMSE of the trained models in the validation set. Then,
we report the RMSE of the best hyperparameters on the test set. Again, we repeat this process
10 times and report the average RMSE, in order to mitigate factors such as randomness in
parameter initialization.
For the regression tree, the only hyperparameter is the maximum number of leave nodes.
The random forest also considers this parameter and also the number of trees. Finally, for the
neural network we tune the following hyperparameters: the learning rate, the batch size, the
number of epochs, and specific combinations of number of units and layers. Note that each
setting of the hyperparameters determines the size of the model. For example, in a neural
network, each unit has a number of coefficients equal to the number of units in the previous
layer plus one. In the case of trees, each intermediate node has two parameters (the index of
the variable and the threshold value), and the leaf nodes have one parameter (the value of the
regression). We want to make sure the models are comparable to the prototype-based model
in terms of the number of parameters. Therefore, when doing hyperparameter optimization,
we restrict them to the ones that yield models (roughly) comparable in size to our prototype-
based model (the model with M = 50 for the 12 input variables and learned scales σm has
1250 parameters).
Table 6 shows the RMSE of the different methods on the test set, the number of parameters
of the models, and a column reminding whether the model is explainable.
In the case of the decision tree, by setting the maximum number of leaf nodes to 250, we
get a tree with 1748 parameters, and the test RMSE is 67.6 $/ft2 . In the case of the random
forest, setting max leaf nodes to 50 and number of estimators to 5 yields 1740 parameters and
a test RMSE of 65.8 $/ft2 (in the case of random forest, the error can be further reduced to 64.6

7 Note this refers to a standard implementation of a neural network, not to the custom neural network discussed
in 3.4.

123
306 Annals of Operations Research (2025) 344:287–311

$/ft2 , but using 50 estimators, which yields 17,400 parameters, many orders of magnitude
higher than for our model). The best performing neural network has 3 hidden layers, with 22
units in each layer (amounting to 1321 parameters) and yields a test RMSE of 61.9 $/ft2 .
When we compare these values with the RMSE reported by the prototype-based model
in the previous section (included in the table as a reminder), we verify that its error is
significantly lower than for random forests and regression trees, and on par with the neural
network. We conclude our model is comparable or better to the other alternatives, but with
the added advantages, such as explainability.

4.6 Results with the neural network extension

Finally, we qualitatively evaluate the extension of the model as a neural network introduced
in Sect. 3.4. Specifically, we illustrate the capability to apply the prototype based model to a
so-called an embedding computed from a pre-existing neural network, instead from the input
variables. As discussed in Sect. 3.4, this increases the complexity of the model but hinders
explainability.
For the pre-existing neural network, we use the network trained in the last section, remove
the output layer and add a 2-unit layer. This produces an intermediate representation which,
consistently with Sect. 3.4, we denote as h. Then, we feed h to the prototype-based model.
Implementation-wise, we proceed in two separate steps: first, we transform the properties
xn to embeddings hn by computing a forward pass, and use these embeddings hn to train the
prototype-based model.
Because hn has two dimensions, we can visualize the prototype distribution in this space.
The result is in Fig. 10.
We visualize the prototype modes (red) on top of a sample of properties in the projected
space h. While it may not be possible to assign an interpretable meaning to each projection
component, we identify some structure, e.g. top left is lower prices, bottom right higher
prices. Also, by inspecting prototypes we can still assign a meaning to some of them, as
indicated in the figure, based on the value (such as “expensive apartments with no basement
and good views”).
This experiment is to demonstrate qualitatively the potential extensions, but it suggests
interesting directions for application and research.

5 Discussion

The main desideratum of the prototype-based model is to predict real estate prices mimicking
the process of comparison-based valuation. The results in the previous section have validated
that the model has the correct behaviour, in two aspects. First, the interpretation of the
prototype distributions is as expected: qualitative analysis using data visualizations show
that the prototypes are well-spread across all the space of input variables, as summarized
in Fig. 8. As a matter of fact, when using spatial variables only (latitude and longitude) the
prototypes spread well across all the different physical regions, as depicted in Figs. 4 and 5.
Inspection of the prototype values also indicates a reasonable segmentation of the space and
seemingly well-approximated price estimates.
Second, on top of this qualitative examination, the quantitative evaluation confirms that
the prototypes and values obtained by the model, in different settings, are appropriate. The
strongest argument for this is that the predictive errors obtained by the prototype-based model

123
Annals of Operations Research (2025) 344:287–311 307

are smaller than (i) alternative ways of constructing the prototypes, as summarized in Table
3, and (ii) alternative machine learning methods, as summarized in Table 6.
The experiments have focused on predicting the price per square foot, as this is a well-
understood measure of unit of constructed land. We conducted experiments in two settings:
predicting the price based only on location, and predicting based on location and other vari-
ables describing the property. The first setting is compatible well-acknowledged assumption
in the real estate sector that location is the crucial factor of land price (Gabrielli et al., 2023;
Atack and Margo, 1998). Even though this setting could be regarded as simplistic, a result
like the one in Fig. 4 could be already interesting and actionable by firms with high stakes in
the real estate market: it can be interpreted as a zonal segmentation of the real estate market,
sometimes also referred to as a sub-market analysis (Goodman and Thibodeau, 2007) (as
analyzed in Sect. 4.3). Effectively, this information alone already produces accurate predic-
tions: by inspecting Tables 3 and 6, we actually note that the error of the prototype-based
model using location-only variables is on par with those of machine learning models using
all the variables. We conclude that location information alone, when coupled with a model
that exploits this information adequately, is a very powerful signal.
Quantifying the location cliché
We could actually quantify the “location, location, location” cliche according to our data
and model, which can be evaluated by comparing the R 2 = 1 − RSS/T SS values of our
regression models, where RSS denotes the residual sum-of-squares of the model under
consideration, and T SS denotes the sum-of-squares with respect to the mean of the test data.
For the model using the location variables only, R 2 = 0.56, for the model using all the
variables but constant σ , R 2 = 0.64, and for the model using all the variables, R 2 = 0.73. By
inspecting the RSS values, the second model reduces the variance by about 20% compared
to the first, and the third model reduces the variance by about 34% compared to the first. We
see a big part of the variance is captured by the location variables, but adding more variables
produces a significant reduction. According to this analysis, location is far from being the
only factor that is significant.
However, these ratios also show an apparent limitation of the model, as the fraction of
variance explained by the best model is R 2 = 0.73, indicating that there is a substantial portion
of the variance not accounted for by the model, which can be attributed to unobserved variables
or inherent irreducible noise within the dataset. This is not surprising given the complexity in
the real estate market and is comparable to other studies (Tchuente and Nyawa, 2022). Some
studies have indeed tried to cover this gap by capturing non-conventional variables such as
the quality of the view from the window (Potrawa and Tetereva, 2022) using computer vision
techniques.
In any case, this limitation points the need for caution when utilizing the model for
individual sample predictions, as it may not adequately capture the intricacies of a specific
data point. However, it is essential to recognize the utility of the model when used for group
predictions or to summarize a real estate dataset with a reduced set of prototypes.
Modeling location in a principled way
Another aspect of the model is that it treats location in a general purpose way. Usually,
location is difficult to model. Some previous studies added variables to the model such as
the distance to different employment centers (Ottensmann et al., 2008). Those studies were
focused on inference and precisely studied the effects of proximity to those key locations in
the price. However, when the goal is price prediction with a general-purpose methodology
that is valid across different markets, it will be cumbersome to specify the key locations for

123
308 Annals of Operations Research (2025) 344:287–311

each new study. The prototype-based model will capture the price changes in different areas
regardless of the reason, without having to define specific variables.
Prototype distributions
The proposed model employs Normal distributions for the prototypes, for several reasons.
First, we note previous prototype-based models, such as Li et al. (2018), work directly with
the L2-norm between prototypes and samples. Therefore, using a distribution such as the
Normal distribution represents a principled generalization. This has several advantages. First,
as demonstrated in Table 5, using a distribution allows fitting additional parameters (σ ’s) and
this has a beneficial impact in reducing the prediction error. Still, as discussed in Sect. 3.3,
the significance of a prototype p in predicting a property x reduces to the Euclidean (or scaled
Euclidean) distance between p and x when the covariance is diagonal. A second reason is that
our keras implementation leverages the tensorflow_probability Python module,
which supports Normal distributions. All in all, the choice of distribution aligns with common
practice. However, it is important to note that our model is not limited to Normal distributions
and can accommodate other types of unimodal distributions for the prototypes.
Explainability
Finally, it is also worth pointing this method could be considered a XAI method, because
the prototypes and values can be thought of as a summary of the dataset and guide the
explanation of predictions. This was already noted in a computer vision task (Li et al., 2018).
While in real estate studies XAI techniques have also been used, we are not aware of previous
prototype-based models—which are especially attractive for this scenario.

6 Practical and research implications

While the text has made reference to various practical implications of the model, they can
be explicitly summarized as follows. First, the proposed model allows firms to perform
a more accurate prediction of real estate properties. Consequently, firms can make better-
informed investment decisions, optimize pricing strategies, and improve customer satisfaction
by providing more accurate and understandable property valuations. However, where the
model excels is in providing explanations in the forms of direct comparisons. Users of such
a model can easily understand the rationale for a predicted price, on the basis of comparing
it with a nearby “prototype”, which could drastically change the relation in which real estate
agents, and other stakeholders, interact with ML systems. Moreover, the prototypes can be
used as “new data” to enrich operations of certain firms.
Beyond industry players, the model can also benefit governments and social actors by
providing insights into the factors driving price dynamics. This understanding can aid in the
development of social impact initiatives and policies aimed at addressing housing affordabil-
ity and other real estate-related social issues. Thus, the model not only promises significant
commercial benefits but also holds potential for broader societal impact.
From a research standpoint, this work introduces novel perspectives on interpreting
machine learning models for real estate predictions. Traditionally, explainable artificial intel-
ligence (XAI) approaches have concentrated on determining the relevance of each variable, as
seen in studies by Lorenz et al. (2022) and Potrawa and Tetereva (2022). Even recent works,
such as those by Bauer et al. (2023) and Holstein et al. (2023), have explored how real estate
professionals utilize model information, still with a primary focus on feature importance but
lacking other cues such as proximity or similarity, used traditionally by real estate agents. The

123
Annals of Operations Research (2025) 344:287–311 309

present work unveils this gap and demands for more intuitive and human-centric approaches
in the real estate price prediction, with the aim of contributing to the transformation of this
research field.

7 Conclusions and future perspectives

This article has introduced the application of prototype-based models in real estate valuation.
Our research started with four desiderata that a model should ideally exhibit for real estate
estimation: (a) non-linearity, (b) parametric learning using a loss function, (c) explainability,
and (d) ability to implement the “direct comparison” criterion. Experimental results sup-
port that the proposed model has the four characteristics simultaneously, which was not the
case of other machine learning models. Moreover, this is attained without sacrificing predic-
tive accuracy: experimental results indicate that the prediction error obtained by the model
is significantly lower than decision trees, random forests or KNN, and on par with neural
networks (remarking, again, that the latter does not comply with the four characteristics dis-
cussed above). We also tried to highlight the usefulness of some of these characteristics, such
as the capability to summarize a dataset with a reduced set of prototypes and the possibility
to draw these prototypes on a map. We hope the method will be useful to operations research
academics and practitioners which have shown previous interest in real estate valuation, as
well as for the specific operations of some firms.
Future lines of research include using more complex form of the prototypes and values. In
the present study, the prototypes are Normal distributions and the values are constant and we
could explore other distributions or non-constant values. Perhaps if we could perform variable
selection within the model we would obtain an even higher degree of explainability. Finally,
we also acknowledge that there could be some use cases where researchers or practitioners are
limited to the use of linear models (e.g. when they perform inference or work in an industry
constrained by heavy regulation, such as banking or insurance). In such a situation, we could
investigate if we could use the prototype-based model as a “feature engineering” method,
where for example the obtained values of γ (Eq. 2) are used as additional location-related
variables in the linear models.
Funding Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence,
and indicate if changes were made. The images or other third party material in this article are included in the
article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is
not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

References
Alexandridis, A. K., Karlis, D., Papastamos, D., & Andritsos, D. (2019). Real estate valuation and forecasting in
non-homogeneous markets: A case study in Greece during the financial crisis. Journal of the Operational
Research Society, 70(10), 1769–1783.
Amédée-Manesme, C. O., & Barthélémy, F. (2018). Ex-ante real estate value at risk calculation method. Annals
of Operations Research, 262, 257–285.

123
310 Annals of Operations Research (2025) 344:287–311

Atack, J., & Margo, R. A. (1998). Location, location, location! the price gradient for vacant urban land: New
York, 1835 to 1900. The Journal of Real Estate Finance and Economics, 16(2), 151–172.
Bauer, K., von Zahn, M., & Hinz, O. (2023). Expl (AI) ned: The impact of explainable artificial intelligence
on users’ information processing. Information systems research, 34(4), 1582–1602.
Baydin, A. G., Pearlmutter, B. A., Radul, A. A., & Siskind, J. M. (2018). Automatic differentiation in machine
learning: A survey. Journal of Machine Learning Research, 18, 1–43.
Biehl, M., Hammer, B., & Villmann, T. (2016). Prototype-based models in machine learning. Wiley
Interdisciplinary Reviews: Cognitive Science, 7(2), 92–111.
Bishop, C. M. (2006). Pattern recognition and machine learning. Berlin: Springer.
Bitter, C., Mulligan, G. F., & Dall’erba, S. (2007). Incorporating spatial variation in housing attribute prices:
A comparison of geographically weighted regression and the spatial expansion method. Journal of
Geographical Systems, 9, 7–27.
Borst, R. A., & McCluskey, W. J. (2008). Using geographically weighted regression to detect housing
submarkets: Modeling large-scale spatial variations in value. Journal of Property Tax Assessment &
Administration, 5(1), 21–54.
Breiman, L. (2001). Random forests. Machine Learning, 45, 5–32.
Chen, S., & Xie, W. (2022). On cluster-aware supervised learning: Frameworks, convergent algorithms, and
applications. INFORMS Journal on Computing, 34(1), 481–502.
Choy, L. H., & Ho, W. K. (2023). The use of machine learning in real estate research. Land, 12(4), 740.
Colwell, P. F., & Dilmore, G. (1999). Who was first? An examination of an early hedonic study. Land
Economics, 620–626.
Court, A. (1939). The dynamics of automobile demand. General Motors: Chapter Hedonic price indexes with
automotive examples.
d’Amato, M., & Kauko, T. (2017). Advances in automated valuation modeling. Springer International
Publishing AG. 10: 978-3.
Doumpos, M., Papastamos, D., Andritsos, D., & Zopounidis, C. (2021). Developing automated valuation
models for estimating property values: A comparison of global and locally weighted approaches. Annals
of Operations Research, 306, 415–433.
Fotheringham, A. S., Brunsdon, C., & Charlton, M. (2003). Geographically weighted regression: The analysis
of spatially varying relationships. Wiley.
Gabrielli, L., Ruggeri, A. G., & Scarpa, M. (2023). Location, location, location: Fluctuations in real estate
market values after covid-19 and the war in Ukraine based on econometric and spatial analysis, random
forest, and multivariate regression. Land, 12(6), 1248.
Gibbs, C., Guttentag, D., Gretzel, U., Morton, J., & Goodwill, A. (2018). Pricing in the sharing economy: A
hedonic pricing model applied to airbnb listings. Journal of Travel & Tourism Marketing, 35(1), 46–56.
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.
Goodman, A. C. (1998). Andrew court and the invention of hedonic price analysis. Journal of Urban
Economics, 44(2), 291–298.
Goodman, A. C., & Thibodeau, T. G. (2007). The spatial proximity of metropolitan area housing submarkets.
Real Estate Economics, 35(2), 209–232.
Haas, G. C. (1922). A statistical analysis of farm sales in blue earth county, Minnesota, as a basis for farm
land appraisal. Technical report.
Holstein, K., De-Arteaga, M., Tumati, L., & Cheng, Y. (2023). Toward supporting perceptual complementarity
in human-ai collaboration via reflection on unobservables. Proceedings of the ACM on Human-Computer
Interaction, 7(CSCW1), 1–20.
Hong, J., Choi, H., & Kim, W. S. (2020). A house price valuation based on the random forest approach:
The mass appraisal of residential property in South Korea. International Journal of Strategic Property
Management, 24(3), 140–152.
Hussain, T., Abbas, J., Wei, Z., & Nurunnabi, M. (2019). The effect of sustainable urban planning and slum
disamenity on the value of neighboring residential property: Application of the hedonic pricing model
in rent price appraisal. Sustainability 11(4). https://doi.org/10.3390/su11041144 .
Isakson, H. R. (1988). Valuation analysis of commercial real estate using the nearest neighbors appraisal
technique. Growth and Change, 19(2), 11–24.
James, G., Witten, D., Hastie, T., Tibshirani, R., et al. (2013). An introduction to statistical learning, (vol.
112). Springer.
Kaden, M., Bohnsack, K. S., Weber, M., Kudła, M., Gutowska, K., Blazewicz, J., & Villmann, T. (2022).
Learning vector quantization as an interpretable classifier for the detection of sars-cov-2 types based on
their rna sequences. Neural Computing and Applications, 34(1), 67–78.
Kok, N., Koponen, E. L., & Martínez-Barbosa, C. A. (2017). Big data in real estate? From manual appraisal
to automated valuation. The Journal of Portfolio Management, 43(6), 202–211.

123
Annals of Operations Research (2025) 344:287–311 311

Kramer, O., & Kramer, O. (2013). K-nearest neighbors. Dimensionality reduction with unsupervised nearest
neighbors, pp. 13–23 .
Li, O., Liu, H., Chen, C., & Rudin, C. (2018). Deep learning for case-based reasoning through prototypes:
A neural network that explains its predictions. In Proceedings of the AAAI conference on artificial
intelligence, vol. 32.
Lins, M. P. E., Novaes, L. F. D. L., Legey, L. F. L. (2005). Real estate appraisal: A double perspective data
envelopment analysis approach. Annals of Operations Research, 138, 79–96.
Lorenz, F., Willwersch, J., Cajias, M., & Fuerst, F. (2022). Interpretable machine learning for real estate market
analysis. Real Estate Economics .
Lundberg, S. M., Erion, G., Chen, H., DeGrave, A., Prutkin, J. M., Nair, B., Katz, R., Himmelfarb, J., Bansal,
N., & Lee, S. I. (2020). From local explanations to global understanding with explainable AI for trees.
Nature Machine Intelligence, 2(1), 56–67.
Malpezzi, S., et al. (2003). Hedonic pricing models: A selective and applied review. Housing Economics and
Public Policy, 1, 67–89.
McCluskey, W. J., Zulkarnain Daud, D., & Kamarudin, N. (2014). Boosted regression trees: An application
for the mass appraisal of residential property in Malaysia. Journal of Financial Management of Property
and Construction, 19(2), 152–167.
McCluskey, W. J., McCord, M., Davis, P. T., Haran, M., & McIlhatton, D. (2013). Prediction accuracy in mass
appraisal: A comparison of modern approaches. Journal of Property Research, 30(4), 239–265.
Ottensmann, J. R., Payton, S., & Man, J. (2008). Urban location and housing prices within a hedonic model.
Journal of Regional Analysis and Policy, 38(1) .
Owomugisha, G., Melchert, F., Mwebaze, E., Quinn, J. A., & Biehl, M. (2021). Matrix relevance learning
from spectral data for diagnosing cassava diseases. IEEE Access, 9, 83355–83363.
Pagourtzi, E., Assimakopoulos, V., Hatzichristos, T., & French, N. (2003). Real estate appraisal: A review of
valuation methods. Journal of Property Investment & Finance, 21(4), 383–401.
Peterson, S., & Flanagan, A. (2009). Neural network hedonic pricing models in mass real estate appraisal.
Journal of Real Estate Research, 31(2), 147–164.
Potrawa, T., & Tetereva, A. (2022). How much is the view from the window worth? machine learning-driven
hedonic pricing model of the real estate market. Journal of Business Research, 144, 50–65.
Rey-Blanco, D., Arbués, P., López, F. A., & Páez, A. (2024). Using machine learning to identify spatial
market segments. A reproducible study of major Spanish markets. Environment and Planning B: Urban
Analytics and City Science, 51(1): 89–108 .
Tchuente, D., & Nyawa, S. (2022). Real estate price estimation in French cities using geocoding and machine
learning. Annals of Operations Research, 1–38 .
Tchuente, D., Lonlac, J., & Kamsu-Foguem, B. (2024). A methodological and theoretical framework for
implementing explainable artificial intelligence (xai) in business applications. Computers in Industry,
155, 104044.
Thibodeau, T. G. (2003). Marking single-family property values to market. Real Estate Economics, 31(1),
1–22.
van Veen, R., Meles, S. K., Renken, R. J., Reesink, F. E., Oertel, W. H., Janzen, A., De Vries, G. J., Leenders,
K. L., & Biehl, M. (2022). Fdg-pet combined with learning vector quantization allows classification of
neurodegenerative diseases and reveals the trajectory of idiopathic rem sleep behavior disorder. Computer
Methods and Programs in Biomedicine, 225, 107042.
Zurada, J., Levitan, A., & Guan, J. (2011). A comparison of regression and artificial intelligence methods in
a mass appraisal context. Journal of Real Estate Research, 33(3), 349–388.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

123