BUSCADOR RECOLECTA

Having in mind the huge amount of data daily registered in the world, it is becoming increasingly important to summarize the information included in a data set. In Statistics and Computer Science, this task is successfully carried out by aggregation functions. One of the most widely applied methodologies of aggregating data is the Choquet integral. The main aim of this paper is to introduce an appropriate notion of Choquet integral in the context of fuzzy numbers. To do this, we face three challenges: the underlying uncertainty when handling fuzzy numbers, the way to order fuzzy numbers by appropriate binary relations and the way to compute the dissimilarity among fuzzy numbers. Illustrative examples are given by involving the α-order on the family of all triangular fuzzy numbers with support on [0,1]., The authors acknowledge with thanks to their universities. A.F. Roldán López de Hierro and H. Bustince are grateful to Ministerio de Ciencia e Innovación by Projects PID2020-119478GB-I00 and PID2022-136627NB-I00, supported by MCIN/AEI/10.13039/501100011033/FEDER, UE. The rest author is also grateful to FEDER 2021-2027/Junta de Andalucía Consejería de Universidad, Investigación e Innovación, project C-EXP-153-UGR23.

Denoising Diffusion Probabilistic Models (DDPMs) have demonstrated exceptional potential across various generative modeling tasks. Despite evident promise in semantic segmentation, their adoption for remote sensing remains limited primarily due to computationally demanding inference. While initial approaches using DDPMs in remote sensing achieve competitive accuracy with state-of-the-art models, the multi-step nature of their image generation process poses a major bottleneck. To address this limitation, this paper investigates three key strategies for accelerating inference: (1) optimizing training and inference steps, (2) applying DDPM acceleration techniques adapted to segmentation task (including Denoising Diffusion Implicit Models, Improved Denoising Diffusion Models, and Progressive Distillation), and (3) thoroughly analyzing the trade-off between accuracy improvement and additional inference time when using test-time augmentation. These strategies are extensively tested with two established remote sensing semantic segmentation datasets focused on buildings and roads. Finally, we compare the optimized diffusion-based model with state-of-the-art convolutional-based models in terms of accuracy and inference times, showing the narrowing gap between both approaches and the increasing viability of diffusion-based segmentation for practical applications., This research received support from an FPU grant (Formación de Profesorado Universitario) awarded to Ruben Pascual (FPU22/02961) by the Spanish Ministry of Science, Innovation and Universities (MICIU). This study was partially funded by the Spanish Ministry of Science and Innovation (MCIN) through the project PID2022-136627NB-I00 (MCIN/AEI/10.13039/501100011033/FEDER, UE).

El reconocimiento de emociones a partir de expresiones faciales (FER) es un campo de investigación importante para la interacción persona-máquina. Sin embargo, los
conjuntos de datos utilizados para entrenar modelos FER a
menudo contienen sesgos demográficos que pueden conducir a la
discriminación en el modelo final. En este trabajo, presentamos
el diseño y la captura realizados para la creación de una nueva
base de datos para FER, donde tratamos de minimizar los sesgos
desde el propio diseño. La base de datos se ha creado utilizando
diferentes métodos de captura. Para comprobar la reducción de
los sesgos alcanzada, analizamos diferentes métricas de sesgo
representacional y estereotípico sobre la base de datos generada
y la comparamos frente a otras bases de datos estándar en la
literatura de FER., Este trabajo ha sido financiado por el MICIU del Gobierno de España
a través del proyecto PID2022-136627NBI00/ AEI/10.13039/501100011033
FEDER, UE y de una ayuda FPU (Rubén Pascual), por el Gobierno de
Navarra (0011-1411-2020-000079 - Emotional Films), y por el Servicio de
Investigación de la Universidad Pública de Navarra (contrato predoctoral de
Iris Domínguez-Catena.)

Convolutional Neural Networks (CNNs) are a family of networks that have become state-of-the-art in several fields of artificial intelligence due to their ability to extract spatial features. In the context of natural language processing, they can be used to build text classification models based on textual features between words. These networks fuse local features to generate global features in their over-time pooling layers. These layers have been traditionally built using the maximum function or other symmetric functions such as the arithmetic mean. It is important to note that the order of input local features is significant (i.e. the symmetry is not an inherent characteristic of the model). While this characteristic is appropriate for image-oriented CNNs, where symmetry might make the network robust to image rigid transformations, it seems counter-productive for text processing, where the order of the words is certainly important. Our proposal is, hence, to use non-symmetric pooling operators to replace the maximum or average functions. Specifically, we propose to perform over-time pooling using pseudo-grouping functions, a family of non-symmetric aggregation operators that generalize the maximum function. We present a construction method for pseudo-grouping functions and apply different examples of this family to over-time pooling layers in text-oriented CNNs. Our proposal is tested on seven different models and six different datasets in the context of engineering applications, e.g. text classification. The results show an overall improvement of the models when using non-symmetric pseudo-grouping functions over the traditional pooling function., The authors acknowledge with thanks to their universities. Furthermore, this work was supported by the Brazilian funding agency
CNPq (Brazilian Research Council) under Projects 311429/2020-3 and
200282/2022-0, by the project PID2022-136627NB-I00 founded by
MCIN/AEI/10.13039/501100011033/FEDER, UE, of the Spanish Government, Project VEGA 1/0193/22 and by Tracasa Instrumental and
the Immigration Policy and Justice Department of the Government of
Navarre.

Overlap and grouping functions can be used to measure events in which we must consider either the maximum or the minimum lack of knowledge. The commutativity of overlap and grouping functions can be dropped out to introduce the notions of pseudo-overlap and pseudo-grouping functions, respectively. These functions can be applied in problems where distinct orders of their arguments yield different values, i.e., in non-symmetric contexts. Intending to reduce the complexity of pseudo-overlap and pseudo-grouping functions, we propose new construction methods for these functions from generalized concepts of additive and multiplicative generators. We investigate the isomorphism between these families of functions. Finally, we apply these functions in an illustrative problem using them in a time series prediction combined model using the IOWA operator to evidence that using these generators and functions implies better performance., This work was supported by the Brazilian funding agency CNPq (Brazilian Research Council) under Projects 311429/2020-3 and 200282/2022-0, and by the project PID2022-136627NB-I00 of the Agencia Estatal de Investigación (Spanish Government), by the grant VEGA 1/0193/22 and by Tracasa Instrumental and the Immigration Policy and Justice Department of the Government of Navarra.

Artificial Intelligence (AI) methods are becoming essential in healthcare. In the context of Intensity-Modulated Radiation Therapy (IMRT), Knowledge-Based Planning (KBP) methodologies have enabled the modification of treatments in real-time to accommodate morphological changes in patients. KBP for IMRT is a data-driven approach that utilises real-time medical imaging to adjust the radiation dose for a patient as needed for the different stages of an illness. In this work we present an interpretable AI model that selects the best IMRT treatment alternatives and determines which is the best. We use an Adaptive Neuforuzzy Adaptive Inference System (ANFIS), which combines the potential of a neural network with the interpretability of a rule based system. We train the model in a supervised manner using the OpenKBP challenge data repository. For this purpose, we also developed a data augmentation method that is supported by Diffusion Probabilistic Models. This approach enables the generation of a wider spectrum of treatment qualities and aids regularisation. The primary advantage of this framework resides in its ability to offer explanations, which is essential in the deployment of medical procedures in real life. Moreover, it serves as a valuable means to test hypotheses concerning the quality of IMRT treatments. Our study reveals that the developed tool has substantial potential to establish itself as a reference in the realm of explainable IMRT treatment selection tools., Xabier Gonzalez s and Humberto Bustince s research has been supported by the project PID2022-136627NB-I00 funded by MCIN/AEI/10.13039/501100011033/FEDER, UE. Javier Fumanal-Idocin s research has been supported by the European Union under a Marie Sklodowska-Curie YUFE4 postdoc action.

Discrete fuzzy integrals (F-integrals) are fusion functions that leverage discrete fuzzy measures to capture interactions within the data. However, their scalability is often limited by the computational overhead of evaluating the measure across the entire measurable space. This paper introduces an efficient online approach for generating fuzzy measures using aggregation functions. The online methodology allows to calculate the F-integral alongside the fuzzy measure without increasing its asymptotic complexity and without requiring previous calculations. The role of the aggregation functions is to establish the properties of the generated measure. To this end, we define and study non-conjunctive aggregation functions, designed to prevent vanishing measures and ensure that the resulting measures retain meaningful and useful properties. In addition the methodology includes an optimizable component, enabling the learning of fuzzy measures and therefore the use of F-Integrals in learning environments. A complexity analysis confirms the method's efficiency, and experiments on supervised classification tasks demonstrate its practical utility., This work was supported in part by Oracle Cloud credits and related resources provided by the Oracle Strategic Partner & Innovation program. Open access funding provided by Universidad Pública de Navarra. Xabier Gonzalez-Garcia's and Humberto Bustince's research has been supported by the PID2022-136627NB-I00 project funded by MCIN/AEI/10.13039/501100011033/FEDER, UE. L'ubomíra Horanská has been supported by the project VEGA 1/0239/24 and VEGA
1/0318/25. Thanks to Dr. Javier Fumanal-Idocín for his support with the Ex-Fuzzy library.

In this paper, the idea of myopic preference and myopic topology provided by Brown and Lewis is further explored. In this sense, new definitions of myopia are introduced for finite spaces. The main contribution of the work is the inclusion of the past in the models. We have defined the notions of past-myopic preference and topology for sequence spaces and n-dimensional spaces. Adding this new dimension makes it possible to work with decision spaces where the economic agent only has information about past events and when she has to choose accordingly to it, which is in line with the reality of certain economic situations, such as voting or finances. This approach generates a wide room for future research lines related to the idea of myopia. Based on this, it would allow to study in forthcoming research past-hyperopic topologies and preferences, and to interconnect different preferences, defined on the past and future models., This work was supported by the project of reference PID2022-136627NB-I00 from MCIN/AEI/10.13039/501100011033/FEDER, UE, by grant PID2021-127799NB-I00 from MCIN/AEI/10.13039/501100011033 and by ERDF A way ofmaking Europe.

Research on the construction of logical connectives using total (admissible) orders is a prolific area of study. Using such orders, a new method for constructing implication functions is defined on the set of discrete fuzzy numbers with support of a closed interval of a given finite chain and whose membership values belong to a finite set of fixed values. This method is based on the use of discrete implication functions defined on a finite chain. Furthermore, a bijective correspondence between the set of implication functions on the aforementioned subset of discrete fuzzy numbers and the set of discrete implication functions defined on the discrete chain is shown. Basic properties of these implication functions are thoroughly investigated, concluding that they are preserved under the proposed construction method. This result highlights the robustness and generality of the method, providing a systematic way to extend discrete implication functions to more complex structures while preserving their underlying properties., This work was partially supported by the R+D+i Projects PID2020-113870GB-I00—“Desarrollo de herramientas de Soft Computing para la Ayuda al Diagnóstico Clínico y a la Gestión de Emergencias (HESOCODICE)” funded by MICIU/ AEI /10.13039/501100011033/ and PID2022-136627NB-I00 funded byMCIN/AEI/10.13039/501100011033/FEDER, UE.

In this work, we propose a new family of aggregation functions inspired by the well-known Choquet integral. To build these functions, we replace the measure in the definition of the Choquet integral by an appropriate function. We study the properties of these aggregation functions and explore the relations with some other common aggregation functions such as order statistics and overlap and grouping functions., H. Bustince, J. Fernandez and G.P. Dimuro are partially supported by research project PID2022-136627NB-I00, Spain ( MCIN/AEI/10.13039/501100011033/ FEDER, UE ). G.P. Dimuro also thanks FAPERGS (Proc. 23/2551-0000126-8 ) and CNPq (Proc. 304118/2023-0 , 407206/2023-0 ). R. Mesiar is supported by Project VEGA 1/0036/23.

Versión original extendida en: https://academica-e.unavarra.es/handle/2454/46368, Este artículo es un resumen del trabajo publicado en la revista Information Fusion [1]. En este artículo explorábamos el reemplazo del operador de pooling máximo comunmente empleado en redes neuronales convolucionales (CNNs) por funciones (a, b)-grouping. Estas funciones extienden el concepto de función de grouping clásica [2] a un intervalo cerrado [a, b], siguiendo la filosofía de [3]. En el contexto del operador de pooling, estas nuevas funciones ayudan a la optimización de los modelos suavizando los gradientes en el proceso de retropropagación y obteniendo resultados competitivos con métodos más complejos, PID2022-136627NB-I00 financiado por MCIN/AEI/10.13039/501100011033/FEDER, UE. Proyecto CNPq (301618/2019-4). Andalusian Excellence project P18-FR-4961. Proyecto VEGA 1/0267/21. Gobierno de Navarra - Departamento de Universidad, Innovación y Transformación Digital. Tracasa Instrumental.

Ordered Weighted Averaging (OWA) operators are some of the most widely used aggregation functions in classic literature, but their application to fuzzy numbers has been limited due to the complexity of defining a total order in fuzzy contexts. However, the recent notion of admissible order for fuzzy numbers provides an effective method to totally order them by refining a given partial order. Therefore, this paper is devoted to defining OWA operators for fuzzy numbers with respect to admissible orders and investigating their properties. Firstly, we define the OWA operators associated with such admissible orders and then we show their main properties. Afterward, an example is presented to illustrate the applicability of these AOWA operators in linguistic decision-making. In this regard, we also develop an admissible order for trapezoidal fuzzy numbers that can be efficiently applied in practice., This study was funded by National Council for Scientific and Technological Development (CNPq-Brazil) within Projects 301618/2019-4, 312053/2018-5, 311429/2020-3, and 200282/2022-0, by Coordination for the Improvement of Higher Education Personnel (CAPES-Brazil) within the project Capes-Print 88887.363001/2019-00, by FAPERGS/Brazil (Proc. 19/2551-0001660-3), by PID2022-136627NB-I00 of the Spanish Government. A.F. Roldán López de Hierro is grateful to Ministerio de Ciencia e Innovación by Project PID2020-119478 GB-I00 and to Junta de Andalucía Program FEDER Andalucía 2014-2020, Project A-FQM-170-UGR20. This work has been also partially supported by the Spanish Ministry of Economy and Competitiveness through the Spanish National Project PGC2018-099402-B-I00, the FEDER-UJA project 1380637, and by the Spanish Ministry of Science, Innovation and Universities through a Formación de Profesorado Universitario grant (FPU2019/01203).; Funding text 2: This study was funded by National Council for Scientific and Technological Development (CNPq-Brazil) within Projects 301618/2019-4 , 312053/2018-5 , 311429/2020-3 , and 200282/2022-0 , by Coordination for the Improvement of Higher Education Personnel (CAPES-Brazil) within the project Capes-Print 88887.363001/2019-00 , by FAPERGS /Brazil (Proc. 19/2551-0001660-3 ), by PID2022-136627NB-I00 of the Spanish Government . A.F. Roldán López de Hierro is grateful to Ministerio de Ciencia e Innovación by Project PID2020-119478 GB-I00 and to Junta de Andalucía Program FEDER Andalucía 2014¿2020, Project A-FQM-170-UGR20 . This work has been also partially supported by the Spanish Ministry of Economy and Competitiveness through the Spanish National Project PGC2018-099402-B-I00 , the FEDER-UJA project 1380637 , and by the Spanish Ministry of Science, Innovation and Universities through a Formación de Profesorado Universitario grant (FPU2019/01203).

Demographic biases in source datasets have been shown as one of the causes of unfairness and discrimination in the predictions of Machine Learning models. One of the most prominent types of demographic bias are statistical imbalances in the representation of demographic groups in the datasets. In this paper, we study the measurement of these biases by reviewing the existing metrics, including those that can be borrowed from other disciplines. We develop a taxonomy for the classification of these metrics, providing a practical guide for the selection of appropriate metrics. To illustrate the utility of our framework, and to further understand the practical characteristics of the metrics, we conduct a case study of 20 datasets used in Facial Emotion Recognition (FER), analyzing the biases present in them. Our experimental results show that many metrics are redundant and that a reduced subset of metrics may be sufficient to measure the amount of demographic bias. The paper provides valuable insights for researchers in AI and related fields to mitigate dataset bias and improve the fairness and accuracy of AI models., This work was supported in part by a predoctoral fellowship from the Research Service of the Universidad Pública de Navarra through open access funding, in part by the Spanish MICIN under Grants PID2019-108392GB-I00, PID2020-118014RB-I00, and PID2022-136627NB-I00/AEI/10.13039/501100011033 FEDER, UE, and in part by the Government of Navarre under Grant 0011-1411-2020-000079 - Emotional Films.

This paper introduces Federated Fuzzy k-Nearest Neighbors algorithms, with one variant designed for classification tasks and another for regression tasks. The federated nature of FedFKNN enables collaborative learning while preserving privacy by keeping sensitive data local to the devices. In our proposal, the central server sends the query to each client. The clients process the query using the k-Nearest Neighbors (k-NN) algorithm locally and return their results to the server. Finally, the server aggregates these results, following the corresponding k-NN algorithm, to produce the final output. Experimental results demonstrate the effectiveness of the federated approach, outperforming the results of a non-collaborative local learning approach., This work is supported by FAPERGS (23/2551-0001865-9, 23/2551-0000126-8), CNPq (304118/2023-0, 407206/2023-0), MCIN/AEI/10.13039/50100011033/FEDER,UE (PID2022-136627NB-I00) and Grant Santander-UPNA 2022.

We deal with two apparently disparate theories. One of them studies extensions of orderings from a set to its power set. The other one defines suitable scores on hesitant fuzzy elements. We show that both theories have the same mathematical substrate. Thus, important possibility/impossibility results concerning criteria for extensions can be transferred to new results on scores. And conversely, conditions imposed a priori on scores can give rise to new extension criteria. This enhances and enriches both theories. We show examples of translations of classical results on extensions in the context of scores. Also, we state new results concerning the impossibility of finding a utility function representing some kind of extension order if some restrictions are imposed on the utility function considered as a score., This work was supported by the project of reference PID2022-136627NB-I00 from MCIN/AEI/10.13039/501100011033/FEDER, UE, and by ERDF A way of making Europe.

Fuzzy Rule-Based Classification System (FRBCS) is a well-known technique to deal with classification problems. Recent studies have considered the usage of the Choquet integral and its generalizations (e.g.: 𝐶𝑇 -integral, 𝐶𝐹 - Integral and 𝐶𝐶-integral) to enhance the performance of such systems. Such fuzzy integrals were applied to the Fuzzy Reasoning Method (FRM) to aggregate the fired fuzzy rules when classifying new data. However, the Sugeno integral, another well-known aggregation operator, obtained good results in other applications, such as brain–computer interfaces. These facts led to the present study, in which we consider the Sugeno integral in classification problems. That is, the Sugeno integral is applied in the FRM of a widely used FRBCS, and its performance is analyzed over 33 different datasets from the literature, also considering different fuzzy measures. To show the efficiency of this new approach, the results obtained are also compared with previous studies that involved the application of different aggregation functions. Finally, we perform a statistical analysis of the application., The authors would like to thank CNPq (proc. 304118/2023-0, 407206/2023-0, 305805/2021-5, 301618/2019-4), FAPERGS (proc. 19/2551-0001660-3, 24/2551-0001396-2), FAPERGS/CNPq (23/2551-0001865-9, 23/2551-0000126-8), Navarra de Servicios y Tecnologías, S.A. (NASERTIC), Grant Santander-UPNA 2021-2022, and PID2022-136627NB-I00 financiado por MCIN/AEI/10.13039/501100011033/FEDER, UE. Open access funding provided by Universidad Pública de Navarra.

In this work, we introduce the notion of grouping index for type-2 fuzzy sets as a measure of how far the union of two type-2 fuzzy sets over the same universe is from the total universe. We also show how we can extend the notion of the Jaccard index to the type-2 setting by means of type-2 grouping and overlap indexes., A.F. Roldán López de Hierro is grateful to FEDER 2021-2027/Junta de Andalucía-Consejería de Universidad, Investigación e Innovación, project C-EXP-153-UGR23. The authors are also grateful to Ministerio de Ciencia e Innovación by Projects PID2020-119478GB-I00 and PID2022-136627NB-I00, supported by MCIN/AEI/10.13039/501100011033/FEDER, UE.

This paper addresses the challenge of generating unlimited new and distinct characters that encompass the style and shared visual characteristics of a limited set of human designed characters. This is a relevant problem in the audiovisual industry, as the ability to rapidly produce original characters that adhere to specific characteristics greatly increases the possibilities in the production of movies, series, or video games. Our solution is built
upon DreamBooth, a widely extended fine-tuning method for text-to-image models. We propose an adaptation focusing on two main challenges: the impracticality of relying on detailed image prompts for character description and the few-shot learning scenario with a limited set of characters available for training. To solve these issues, we introduce additional character-specific tokens to DreamBooth training and remove its class-specific regularization dataset. For an unlimited generation of characters, we propose the usage of random tokens and random embeddings. This proposal is tested on two specialized datasets and the results shows our method¿s capability to produce diverse characters that adhere to a style and visual characteristics. An ablation study to analyze the contributions of the proposed modifications is also developed., This work has been funded by MCIN/AEI/10.13039/501100011033/FEDER, UE, with the project PID2022-136627NB-I00, by the Government of Navarre under the project 0011-1365-2022-000130, and by the Public University of Navarra under the project PJUPNA2023-11377. The Scary and Virus datasets are designed by Freepik. This research received support from an FPU grant (Formación de Profesorado Universitario) awarded by the Spanish Ministry of Science and Innovation (MCINN) to Rubén Pascual.

In recent years, the construction and characterization of certain uninorms on bounded lattices have been exhaustively studied. In this paper, we study the structure of the set of elements incomparable with the neutral element, and provide a taxonomy of different types of bounded lattices. Moreover, we present different methods for constructing uninorms on some of these types of bounded lattices. The presented construction methods extend existing construction methods in the sense that the constructed uninorm may take a more general range of values on the set of elements incomparable with the neutral element., A. G. is supported by Fundacion Caja Navarra. M. G. is supported by the Agencia Estatal de Investigacion of Spain PID2022-136627NB-I00. R. P.-F. is supported by project PID2022-140585NB-I00 funded by MICIU/AEI/10.13039/501100011033 and FEDER/UE. Open access Funding provided by Universidad Publica de Navarra.

The 2019 coronavirus disease outbreak, caused by the severe acute respiratory syndrome type-2 virus (SARS-CoV-2), was declared a pandemic in March 2020. Since its emergence to the present day, this disease has brought multiple countries to the brink of health care collapse during several waves of the disease. One of the most common tests performed on patients is chest x-ray imaging. These images show the severity of the patient's illness and whether it is indeed covid or another type of pneumonia. Automated assessment of this type of imaging could alleviate the time required for physicians to treat and diagnose each patient. To this end, in this paper we propose the use of Convolutional Neural Networks (CNNs) to carry out this process. The aim of this paper is twofold. Firstly, we present a pipeline adapted to this problem, covering all steps from the preprocessing of the datasets to the generation of classification models based on CNNs. Secondly, we have focused our study on the modification of the information fusion processes of this type of architectures, in the pooling layers. We propose a number of aggregation theory functions that are suitable to replace classical processes and have shown their benefits in past applications, and study their performance in the context of the x-ray classification problem. We find that replacing the feature reduction processes of CNNs leads to drastically different behaviours of the final model, which can be beneficial when prioritizing certain metrics such as precision or recall., The authors gratefully acknowledge the financial support of Tracasa Instrumental (iTRACASA), Spain and of the Gobierno de Navarra - Departamento de Universidad, Innovación y Transformación Digital, Spain, as well as that of the Spanish Ministry of Science (project PID2022-136627NB-I00) and the project PC095-096 FUSIPROD. Z. Takáč is supported by grant VEGA, Slovak Republic 1/0267/21. Open access funding provided by Universidad Pública de Navarra.

Choosing the proper ground sampling distance (GSD) is a vital decision in remote sensing, which can determine the success or failure of a project. Higher resolutions may be more suitable for accurately detecting objects, but they also come with higher costs and require more computing power. Semantic segmentation is a common task in remote sensing where GSD plays a crucial role. In semantic segmentation, each pixel of an image is classified into a predefined set of classes, resulting in a semantic segmentation map. However, comparing the results of semantic segmentation at different GSDs is not straightforward. Unlike scene classification and object detection tasks, which are evaluated at scene and object level, respectively, semantic segmentation is typically evaluated at pixel level. This makes it difficult to match elements across different GSDs, resulting in a range of methods for computing metrics, some of which may not be adequate. For this reason, the purpose of this work is to set out a clear set of guidelines for fairly comparing semantic segmentation results obtained at various spatial resolutions. Additionally, we propose to complement the commonly used scene-based pixel-wise metrics with region-based pixel-wise metrics, allowing for a more detailed analysis of the model performance. The set of guidelines together with the proposed region-based metrics are illustrated with building and swimming pool detection problems. The experimental study demonstrates that by following the proposed guidelines and the proposed region-based pixel-wise metrics, it is possible to fairly compare segmentation maps at different spatial resolutions and gain a better understanding of the model's performance. To promote the usage of these guidelines and ease the computation of the new region-based metrics, we create the seg-eval Python library and make it publicly available at https://github.com/itracasa/ seg-eval., The work of Christian Ayala was supported in part by the Government of Navarre through the Industrial Ph.D. Program 2020 Reference under Grant 0011-1408-2020-000008. The work of Mikel Galar was supported in part by the Spanish Ministry of Science and Innovation (MCIN/Agencia Estatal de Investigación
(AEI)/10.13039/501100011033) under Project PID2019-108392GB-I00 and Project PID2022-136627NB-I00 and in part by the Public University of Navarre under Project PJUPNA25-2022.

El cambio climático afecta a las condiciones medioambientales de las distintas regiones. La capacidad de constatar estos cambios es una eficaz herramienta para adaptarse a la evolución de las condiciones. Los datos meteorológicos se generan continuamente en múltiples estaciones de todo el mundo, proporcionando una valiosa información sobre la variabilidad en el tiempo de los patrones climáticos. El estudio de este flujo de datos nos permite comprender mejor los nuevos patrones climáticos. Este trabajo explora, mediante un algoritmo de agrupamiento de flujos de datos (stream clustering), el potencial de emplear datos meteorológicos obtenidos en diferentes localizaciones geográficas para rastrear el cambio en los patrones climáticos en la Comunidad Foral de Navarra durante los últimos 20 años. El estudio de caso mostró la aplicabilidad de los métodos de flujos de datos a la segmentación incremental de regiones geográficas en función de sus factores climatológicos., El trabajo está parcialmente financiado por del proyecto PID2022-136627NB-I00 de la Agencia Estatal de Investigación financiado por MCIN/AEI/10.13039/501100011033/FEDER,UE. Graçaliz Pereira agradece a la Agencia de Financiación Brasileña CNPq (Proc. 407206/2023-0, 304118/2023-0) (CNDCT Brasil). Asier Urio-Larrea es beneficiario de los Contratos predoctorales Santander-UPNA 2021.

Machine learning models can inherit biases from their training data, leading to discriminatory or inaccurate predictions. This is particularly concerning with the increasing use of large, unsupervised datasets for training foundational models. Traditionally, demographic biases within these datasets have not been well-understood, limiting our ability to understand how they propagate to the models themselves. To address this issue, this paper investigates the propagation of demographic biases from datasets into machine learning models. We focus on the gender demographic component, analyzing two types of bias: representational and stereotypical. For our analysis, we consider the domain of facial expression recognition (FER), a field known to exhibit biases in most popular datasets. We use Affectnet, one of the largest FER datasets, as our baseline for carefully designing and generating subsets that incorporate varying strengths of both representational and stereotypical bias. Subsequently, we train several models on these biased subsets, evaluating their performance on a common test set to assess the propagation of bias into the models¿ predictions. Our results show that representational bias has a weaker impact than expected. Models exhibit a good generalization ability even in the absence of one gender in the training dataset. Conversely, stereotypical bias has a significantly stronger impact, primarily concentrated on the biased class, although it can also influence predictions for unbiased classes. These results highlight the need for a bias analysis that differentiates between types of bias, which is crucial for the development of effective bias mitigation strategies., This work was funded by a predoctoral fellowship and open access funding from the Research Service of the Universidad Publica de Navarra, the Spanish MICIN (PID2020-118014RB-I00 and PID2022-136627NB-I00, AEI/10.13039/501100011033 FEDER, UE), the Government of Navarre (0011-1411-2020-000079 - Emotional Films), and the support of the 2024 Leonardo Grant for Researchers and Cultural Creators from the BBVA Foundation.

Overlap and grouping indices are functions measuring, respectively, the fuzzy intersection and fuzzy union of two fuzzy sets. They have been applied successfully in several fields, such as in interpolative fuzzy systems, fuzzy rule-based classification systems and comparison of fuzzy inference rules. Overlap and grouping indices can be built employing overlap and grouping functions, respectively, which are possibly non-associative aggregation functions with features that provide good results when applied to practical bivariate problems. Many studies have generalized the concepts of overlap and grouping functions to be applied in n-dimensional problems. However, the concepts of overlap/grouping indices have not been generalized in similar pattern. Since the associative property may not hold, their application in n-dimensional domains, for comparing more than two fuzzy sets at a time, is not immediate, which limit their application in such contexts. The objective of this paper is to introduce the concepts of n-dimensional and general overlap/grouping indices, with special attention to the development of their construction methods based on generalized overlap/grouping functions. As an application example, we introduce the concept of n-dimensional Jaccard index, with a construction method based on n-dimensional overlap/grouping indices, providing an n-dimensional fuzzy set similarity score., Open Access funding provided by Universidad Pública de Navarra. Partially funded by: FAPERGS (24/2551-0001396-2, 24/2551-0000723-7, 23/2551-0001865-9), CNPq (304118/2023-0, 407206/2023-0), CAPES, FAPESP (2022/09136-1), FAPERGS/CNPq (23/2551-0000126-8), MCIN/AEI/10.13039/501-00011033/ FEDER,UE (PID2022-136627NB-I00).

Este artículo aborda el reto de generar un número
ilimitado de personajes nuevos, y distintos, que engloben el estilo
y las características visuales compartidas de un conjunto limitado
de personajes diseñados por un humano. Este es un problema de
gran relevancia en la industria audiovisual, ya que la capacidad
de producir rápidamente personajes originales que se adhieran
a unas características específicas aumenta enormemente las
posibilidades en la producción de películas, series o videojuegos.
Nuestra solución se basa en DreamBooth, un método de ajuste de
modelos generativos de texto a imagen ampliamente extendido.
Proponemos una adaptación centrada en dos retos principales:
lo poco práctico que resulta utilizar prompts detallados de las
imágenes para describir los personajes y la complejidad del ajuste
de modelos a partir de un conjunto limitado de personajes. Para
resolver estos problemas, introducimos en el entrenamiento de
DreamBooth tokens adicionales específicos para cada personaje
y eliminamos el conjunto de datos de regularización. Para
generar personajes de manera ilimitada, proponemos el uso
de tokens y embeddings aleatorios. Comprobamos la utilidad
de la propuesta utilizando dos conjuntos de datos diferentes.
Los resultados obtenidos muestran la capacidad de nuestro
método para producir personajes diversos que se adhieren a
un estilo y a unas características visuales concretas. Finalmente,
desarrollamos un estudio de ablación., Este trabajo está financiado por MCIN/AEI/10.13039/501100011033/FEDER, UE, con el proyecto PID2022-136627NB-I00, por el Gobierno de Navarra bajo el proyecto 0011-1365-2022-000130, y por la Universidad Pública de Navarra bajo el proyecto PJUPNA2023-11377. Los conjuntos de datos Scary y Virus están diseñados por Freepik. Esta investigación ha contado con el apoyo de una beca FPU concedida por el Ministerio de Ciencia e Innovación de España (MCINN) a Rubén Pascual.

At present, (quasi-)overlap functions have been extended to various universes of discourse and become a hot research topic. Meanwhile, the investigation of extended aggregation operations for normal convex fuzzy truth values has also attracted much attention. This paper mainly studies the representation of quasi-overlap functions for normal convex fuzzy truth values based on generalized extended overlap functions, which is the fundamental problem in the whole study of overlap functions for normal convex fuzzy truth values. Firstly, we present the definitions of (restrictive-)quasi-overlap functions and lattice-ordered-(restrictive-)quasi-overlap functions for normal convex fuzzy truth values and generalized extended overlap functions, respectively. Secondly, we present the (equivalent) characterizations for the closure properties of generalized extended overlap functions for various fuzzy truth values. Thirdly, we characterize the basic properties of generalized extended overlap functions for normal convex fuzzy truth values. Finally, by an equivalent characterization with a prerequisite, we successfully represent quasi-overlap functions for normal convex fuzzy truth values based on generalized extended overlap functions. Notably, we can quickly obtain (restrictive-)quasi-overlap functions for normal convex fuzzy truth values using the left-continuous quasi-overlap functions on interval [0,1]. Moreover, regarding the relationships between four types of quasi-overlap functions for normal convex fuzzy truth values, the details implication relations are that lattice-ordered-(restrictive-)quasi-overlap functions are strictly stronger than (restrictive-)quasi-overlap functions for normal convex fuzzy truth values even if all of them are constructed by generalized extended overlap functions., Supported by National Natural Science Foundation of China (62166037, 12201512), Science and Technology Program of Gansu Province (20JR10RA101), Longyuan Youth Innovation and Entrepreneurship Talent Project (6017/20231004), Funds for Innovative Fundamental Research Group Project of Gansu Province (23JRRA684), research project PID2022-136627NB-I00, and MCIN/AEI/10.13039/501100011033/FEDER, UE.

In the last few years, Artificial Intelligence (AI) systems have become increasingly widespread. Unfortunately, these systems can share many biases with human decision-making, including demographic biases. Often, these biases can be traced back to the data used for training, where large uncurated datasets have become the norm. Despite our awareness of these biases, we still lack general tools to detect, quantify, and compare them across different datasets. In this work, we propose DSAP (Demographic Similarity from Auxiliary Profiles), a two-step methodology for comparing the demographic composition of datasets. First, DSAP uses existing demographic estimation models to extract a dataset's demographic profile. Second, it applies a similarity metric to compare the demographic profiles of different datasets. While these individual components are well-known, their joint use for demographic dataset comparison is novel and has not been previously addressed in the literature. This approach allows three key applications: the identification of demographic blind spots and bias issues across datasets, the measurement of demographic bias, and the assessment of demographic shifts over time. DSAP can be used on datasets with or without explicit demographic information, provided that demographic information can be derived from the samples using auxiliary models, such as those for image or voice datasets. To show the usefulness of the proposed methodology, we consider the Facial Expression Recognition task, where demographic bias has previously been found. The three applications are studied over a set of twenty datasets with varying properties. The code is available at https://github.com/irisdominguez/DSAP., This work was funded by a predoctoral fellowship from the Research Service of the Universidad Publica de Navarra, the Spanish MICIN (PID2020-118014RB-I00 and PID2022-136627NB-I00/AEI/10.13039/501100011033 FEDER, UE), and the Government of Navarre (0011-1411-2020-000079 - Emotional Films). Open access funding provided by Universidad Pública de Navarra.

En este trabajo presentamos una nueva clase de funciones de agregación. Para la definición de estas nuevas funciones nos inspiramos en el método de construcción de las integrales Choquet, reemplazando las medidas por funciones adecuadas. Tras discutir la definición de las nuevas funciones, estudiamos algunas de su propiedades básicas y consideramos su relación con otras funciones de agregación utilizadas en la literatura, como los estadísticos de orden o las funciones de overlap y grouping., H. Bustince, X. Gonzalez-Garcia and G.P. Dimuro han sido financiados por el proyecto de investigación PID2022- 136627NB-I00 (MCIN/AEI/10.13039/501100011033/FEDER, UE)

We introduce and study aggregation functions based on extremal values, namely extended (𝑙, 𝑢)- aggregation functions whose outputs only depend on a fixed number 𝑙 of extremal lower input values and a fixed number 𝑢 of extremal upper input values, independently of the arity of the input 𝑛-tuples (𝑛 ≥ 𝑙 + 𝑢). We discuss several general properties of (𝑙, 𝑢)-aggregation functions and we study special (𝑙, 𝑢)-aggregation functions with neutral element, including t-conorms, t-norms and uninorms. We also study (𝑙, 𝑢)-aggregation functions defined by means of integrals with respect to discrete fuzzy measures, as well as (𝑙, 𝑢)-ordered weighted quasi-arithmetic means based on appropriate weighting vectors. We also stress some generalizations based on recently introduced new types of monotonicity. Some possible applications are sketched, too., I. Rodriguez-Martinez and H. Bustince gratefully acknowledge the financial support of the Spanish Ministry of Science, project PID2022-136627NB-I00 financed by MCIN/AEI/10.13039/501100011033/FEDER, UE. R. Mesiar and A. Kolesarová are supported by the grant VEGA 1/0036/23. F. Herrera is financially supported by grant P18-FR4961. I. Rodriguez-Martinez is also partially supported by Tracasa Instrumental (iTRACASA) and Gobierno de Navarra - Departamento de Universidad, Innovación y Transformación Digital. R. Halaš was supported by the IGA grant of Palacký University Olomouc IGAPrF2024011.

This paper contributes to the theoretical literature on decision models where agents may encounter challenges in comparing alternatives. We introduce a characterization of countable Richter–Peleg multi-utility representations, both semicontinuous (upper and lower) and continuous, within preorders that may not be total. The proposed theorems provide a comprehensive mathematical framework, complementing previous results of Alcantud et al. and Bosi on countable multi-utility representations. Our characterizations establish necessary and sufficient conditions through topological properties and constructive methods via indicator functions. Furthermore, we introduce a topological framework aligned with the property of strong local non-satiation and provide a novel theorem containing sufficient conditions for the existence of countable upper semi-continuous multi-utility representations of a preorder. The results demonstrate that preference representations can be achieved using countably many functions rather than uncountable families, with implications for computational tractability and the identification of maximal elements in optimization contexts., This work was supported by the project of reference PID2022-136627NB-I00 from MCIN/AEI/10.13039/501100011033/FEDER, UE, and by grant PID2021-127799NB-I00 from MCIN/AEI/10.13039/501100011033, ERDF A way of making Europe.

Climate change has an effect on the environmental conditions of different regions. Being able to track these changes is a powerful tool for adapting to evolving conditions. Weather data is continuously generated across multiple stations around the world, providing valuable information on climate time-varying patterns. Studying this data stream enables us to understand the new climate patterns better. This paper explores, through a stream clustering algorithm, the potential of employing weather data in different geographical locations to track the change in climate patterns in the Spanish region of Navarre over the last 20 years. The case study showed the applicability of stream methods to the incremental segmentation of geographical regions based on their climatology factors. In this study, we have found that the climate of Navarre is homogenising into the particular climate of southwestern regions, which is expanding. This particular finding may raise concerns about the time-varying impact that climate change is having on Navarre regions, where large parts of its geography can be grouped into a single climate., Thanks to PID2022-136627NB-I00 supported by MCIN/AEI/10.13039/501100011033/FEDER,UE. Gracaliz thanks to CNPq (301618/2019-4,305805/2021-5) and FAPERGS (19/2551-0001660-3) (CNDCT Brasil) and Asier Urio-Larrea thanks to Contratos predoctorales Santander-UPNA 2021

Image smoothing is a low-level image processing task mainly aimed at homogenizing an image, mitigating noise, or improving the visibility of certain image areas. There exist two main strategies for image smoothing. The first strategy is content-unaware image smoothing. This strategy replicates identical smoothing behavior at every region in the image, hence ignoring any local or semi-local properties of the image. The second strategy is content-aware image smoothing, which takes into account the local properties of the image in order to adapt the smoothing behavior. Such adaptation to local image conditions is intended to avoid the blurring of relevant structures (such as ridges, edges, and blobs) in the image. While the former strategy was ubiquitous in the early years of image processing, the last 20 years have seen an ever-increasing use of the latter, fueled by a combination of greater computational capability and more refined mathematical models. In this work, we propose a novel content-aware image smoothing method based on soft (fuzzy) clustering. Our proposal capitalizes on the strengths of soft clustering to produce content-aware smoothing and allows for the direct configuration of the most relevant parameters for the task: the number of distinctive regions in the image and the relative relevance of spatial and tonal information in the smoothing. The proposed method is put to the test on both artificial and real-world images, combining both qualitative and quantitative analyses. We also propose the use of a local homogeneity measure for the quantitative analysis of image smoothing results. We show that the proposed method is not sensitive to centroid initialization and can be used for both artificial and real-world images., This work was supported by the project PID2022-136627NB-I00 of the Agencia Estatal de Investigación (Spanish Government), as well as by the Iberus Program (Project num. 801586, funded by the European Union).

Versión original extendida en: https://academica-e.unavarra.es/handle/2454/45222, Este artículo es un resumen del trabajo publicado en la revista IEEE Transactions on Fuzzy Systems. En este trabajo, presentamos una contribución a la teoría de las Funciones de Equivalencia Restringida (REF), que permite comparar elementos multivaluados. Extendemos el concepto de REF de L a Ln y presentamos una nueva construcción de similitud en Ln. A partir de esta filosofía se construyen medidas de similitud entre multiconjuntos difusos y se presenta un ejemplo aplicado en el contexto de la difusión anisotrópica de imágenes en color., Proyecto PID2022-136627NB-I00 financiado por MCIN/AEI/10.13039/501100011033/FEDER, UE; Proyectos TED2021-131295B-C32, PID2021-123673OB-C31 del
MCIN/AEI/10.13039/501100011033 Proyecto PROMETEO CIPROM/2021/077 Conselleria de Innovación, Universidades, Ciencia y Sociedad Digital - Generalitat Valenciana Ayuda PAID-06-23 del Vicerrectorado de Investigación de la U. Politécnica de València (UPV)

Aggregation operators are unvaluable tools when different pieces of information have to be taken into account with respect to the same object. They allow to obtain a unique outcome when different evaluations are available for the same element/object. In this contribution we assume that the opinions are not given in form of isolated values, but intervals. We depart from two “classical” aggregation functions and define a new operator for aggregating intervals based on the two original operators. We study under what circumstances this new function is well defined and we provide a general characterization for monotonicity. We also study the behaviour of this operator when the departing functions are the most common aggregation operators. We also provide an illustrative example demonstrating the practical application of the theoretical contribution to ensemble deep learning models., Authors would like to thank for the support of the Spanish Ministry of Science and Innovation projects PID2022-139886NB-I00 (S. Diaz-Vazquez, E. Torres-Manzanera, N. Rico, I. Diaz and S. Montes) and Ministerio de Educación y Formación Profesional PID2022-136627NB-I00 (I. Rodriguez-Martinez and H. Bustince).

This paper comprehensively investigates the performance of various strategies for predicting CO2 levels in school classrooms over different time horizons by using data collected through IoT devices. We gathered Indoor Air Quality (IAQ) data from fifteen schools in Navarra, Spain between 10 January and 3 April 2022, with measurements taken at 10-min intervals. Three prediction strategies divided into seven models were trained on the data and compared using statistical tests. The study confirms that simple methodologies are effective for short-term predictions, while Machine Learning (ML)-based models perform better over longer prediction horizons. Furthermore, this study demonstrates the feasibility of using low-cost devices combined with ML models for forecasting, which can help to improve IAQ in sensitive environments such as schools., A.J. and D.P. were partially supported by the Spanish Ministry of Science and Innovation through the project PID2022-136627NB-I00 (MCIN/AEI/10.13039/501100011033/FEDER, UE). P.G.-P. was supported by the Gobernment of Navarra under 'Doctorados Industriales 2021'.

Automatic art analysis employs different image processing techniques to classify and categorize works of art. When working with artistic images, we need to take into account further considerations compared to classical image processing. This is because artistic paintings change drastically depending on the author, the scene depicted, and their artistic style. This can result in features that perform very well in a given task but do not grasp the whole of the visual and symbolic information contained in a painting. In this article, we show how the features obtained from different tasks in artistic image classification are suitable to solve other ones of similar nature. We present different methods to improve the generalization capabilities and performance of artistic classification systems. Furthermore, we propose an explainable artificial intelligence method to map known visual traits of an image with the features used by the deep learning model considering fuzzy rules. These rules show the patterns and variables that are relevant to solve each task and how effective is each of the patterns found. Our results show that compared to multitask learning, our proposed context-aware features can achieve up to 19% more accurate results when using the residual network architecture and 3% when using ConvNeXt. We also show that some of the features used by these models can be more clearly correlated to visual traits in the original image than other kinds of features., This work was supported in part by the Oracle Cloud credits and related resources provided by Oracle, in part by the MCIN/AEI/10.13039/501100011033 and ERDF "A way of making Europe" under Grant CONFIA PID2021-122916NB-I00, and in part by the MCIN Project PID2022-136627NB-I00.

In data stream (DS) learning, the system has to extract knowledge from data generated continuously, usually at high speed and in large volumes, making it impossible to store the entire set of data to be processed in batch mode. Hence, machine learning models must be built incrementally by processing the incoming examples, as data arrive, while updating the model to be compatible with the current data. In fuzzy DS clustering, the model can either absorb incoming data into existing clusters or initiate a new cluster. As the volume of data increases, there is a possibility that the clusters will overlap to the point where it is convenient to merge two or more clusters into one. Then, a cluster comparison measure (CM) should be applied, to decide whether such clusters should be combined, also in an incremental manner. This defines an incremental fusion process based on aggregation functions that can aggregate the incoming inputs without storing all the previous inputs. The objective of this article is to solve the fuzzy DS clustering problem of incrementally comparing fuzzy clusters on a formal basis. First, we formalize and operationalize incremental fusion processes of fuzzy clusters by introducing recursively extendable (RE) aggregation functions, studying construction methods and different classes of such functions. Second, we propose two approaches to compare clusters: 1) similarity and 2) overlapping between clusters, based on RE aggregation functions. Finally, we analyze the effect of those incremental CMs on the online and offline phases of the well-known fuzzy clustering algorithm d-FuzzStream, showing that our new approach outperforms the original algorithm and presents better or comparable performance to other state-of-the-art DS clustering algorithms found in the literature., This work was supported in part by FAPERGS under Grant 24/2551-0001396-2, Grant 23/2551-0001865-9, and Grant 24/2551-0000723-7; in part by CNPq under Grant 304118/2023-0 and
Grant 407206/2023-0; in part by FAPERGS/CNPq under Grant 23/2551- 0000126-8; in part by CAPES, FAPESP under Grant 2022/09136-1; in part by MCIN/AEI/10.13039/50100011033/FEDER; and in part by UE under Grant PID2022-136627NB-I00 and Grant Santander-UPNA.

Zadeh has defined its consistency index OZ in order
to compare two fuzzy subsets A and B of a referential set U,
in the sense that “The lower OZ (A, B) is, the closer to ∅ A ∩ B
is.” Several axiomatizations of this concept can be found in the
literature, in terms of overlap indices, allowing its application in
different contexts, as in classification and clustering. The present
work, in an opposite direction, introduces the grouping index, a
comparison index G of two fuzzy subsets A and B that provides
a measure of the proximity of the fuzzy union of A and B to the
referential set U, that is, ‘the higher G(A, B) is, the closer to U A∪
B is.” Many important properties are analyzed, and construction
methods are presented, whereas establishing the formal relation
between overlap and grouping indices via duality. The theory is
illustrated by several detailed examples., This work was developed in the context of an international cooperation project among the universities UPNA and FURG, and partially supported by PID2022-136627NB-I00 funded by MCIN/AEI/10.13039/501100011033/FEDER, UE, and CNPq (Proc. 407206/2023-0, 304118/2023-0).