Article
Computer Science, Information Systems
Martha Flores-Sosa, Ezequiel Aviles-Ochoa, Jose M. Merigo, Ronald R. Yager
Summary: The research discusses the importance of volatility and proposes an estimator that combines OWA operators with OLS. By incorporating OWA operators into ARCH-GARCH models, a method that can handle high levels of uncertainty is developed, ultimately achieving efficient forecasting in MX/US exchange rate volatility.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
LeSheng Jin, Zhen-Song Chen, Ronald R. Yager, Tapan Senapati, Radko Mesiar, Diego Garcia Zamora, Bapi Dutta, Luis Martinez
Summary: This study introduces a novel approach to establish OWA operators for basic uncertain information, using problem factorization and integration techniques similar to the Choquet integral. It also discusses various methodologies for determining specific weights to define these operators.
INFORMATION SCIENCES
(2023)
Article
Mathematics
Pere Josep Pons-Vives, Mateu Morro-Ribot, Carles Mulet-Forteza, Oscar Valero
Summary: This paper proposes an improved algorithm, OWA-based K-means, for clustering customers based on their spending propensity. Experiments show that the use of OWA operator improves the performance of classical K-means significantly. The OWA-based K-means can be applied to classify customers in different seasons without requiring radical changes in the implementation of the classical method or additional implementation costs in real hotel management.
Article
Computer Science, Artificial Intelligence
Martha Flores-Sosa, Ernesto Leon-Castro, Jose M. Merigo, Ronald R. Yager
Summary: This paper introduces the MLR-HOWA operator, which uses HOWA means to obtain beta values. It provides the possibility of under or overestimating results based on the decision maker's expectations and knowledge, allowing for analysis of multiple scenarios from minimum to maximum. The paper also presents the main properties of the operator and two extensions using induced and generalized variables. An application in exchange rate forecasting for five Latin American countries is provided, demonstrating that using different combinations of MLR with OWA operators can reduce forecasting errors.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Automation & Control Systems
Zhonghao Xie, Xi'an Feng, Xiaojing Chen
Summary: This paper proposes a robust method for PLS based on the idea of least trimmed squares (LTS), which effectively deals with high-dimensional regressors. By formulating the LTS problem as a concave maximization problem, the complexity of solving LTS is simplified. The results from simulation and real data sets demonstrate the effectiveness and robustness of the proposed approach.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Juan Baz, Mikel Ferrero-Jaurrieta, Irene Diaz, Susana Montes, Gleb Beliakov, Humberto Bustince
Summary: This paper studies the aggregation of multiple predictors in time series forecasting and introduces a new pre-aggregation extension operator. By examining the behavior and performance of the operator from a probabilistic perspective, its effectiveness is validated in practical examples.
INFORMATION FUSION
(2024)
Article
Computer Science, Theory & Methods
Radko Mesiar, Andrea Stupnanova, LeSheng Jin
Summary: This paper introduces OWA operators and their representation based on Choquet integrals, and proposes a generalization of OWA operators called BIOWA operators. The properties of BIOWA operators are studied and exemplified through various examples.
FUZZY SETS AND SYSTEMS
(2022)
Article
Environmental Sciences
Jiangyue Li, Xi Chen, Alishir Kurban, Tim Van de Voorde, Philippe De Maeyer, Chi Zhang
Summary: This study quantified the spatiotemporal changes of ecosystem services in the major basins of Central Asia and identified conservation priorities using a multi-criterion valuation method. The results showed variations in conservation efficiency among different basins, with ecosystem services being influenced by both natural and socioeconomic factors.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2021)
Article
Engineering, Electrical & Electronic
Debasis Kundu, Swagata Nandi, Rhythm Grover
Summary: The paper proposes the use of weighted least squares estimators for parameter estimation in chirp models. These estimators are robust to outliers and have consistent and convergent properties. Extensive simulations were performed to validate the performance of the proposed estimators.
CIRCUITS SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Chemistry, Analytical
Maxime Metz, Florent Abdelghafour, Jean-Michel Roger, Matthieu Lesnoff
Summary: The paper presents a novel robust PLSR algorithm, RoBoost-PLSR, inspired by boosting principles, which mitigates the impact of outliers during calibration. Compared to other algorithms, RoBoost-PLSR demonstrates resilience and good performance on multiple datasets.
ANALYTICA CHIMICA ACTA
(2021)
Article
Green & Sustainable Science & Technology
Betzabe Ruiz-Morales, Irma Cristina Espitia-Moreno, Victor G. Alfaro-Garcia, Ernesto Leon-Castro
Summary: This study proposes a new method using OWA operators to analyze the SDGs index. By evaluating the relative importance of each SDG and using OWA and POWA operators, rankings were generated, showing that country rankings can change depending on the weights of each SDG.
Article
Engineering, Electrical & Electronic
Cody Mazza-Anthony, Bogdan Mazoure, Mark Coates
Summary: This paper introduces two novel estimators based on the OWL norm for estimating sparse structured precision matrices, which can simultaneously identify groups of related edges and control sparsity. The ccGOWL estimator shows good computational efficiency and accuracy in both synthetic data and real-world applications, demonstrating its efficacy in gene network analysis and econometrics.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2021)
Article
Computer Science, Information Systems
Thomas Servotte, Jakob Raymaekers, Tim Verdonck
Summary: This study proposes two improvements for the robust regression methods of least trimmed squares (LTS) and least median of squares (LMS). These improvements involve the use of soft differentiable sorting in the loss functions and deterministic initialization of the estimators using the wrapping transformation. Experimental results show that these improvements lead to increased predictive accuracy and faster convergence. Moreover, the soft loss function has a greater impact on the LMS method.
INFORMATION SCIENCES
(2023)
Article
Energy & Fuels
Aihua Tang, Chun Wang, Dongyang Zhang, Kaiqing Zhang, Yapeng Zhou, Zhigang Zhang
Summary: This paper proposes a multi-model SOC fusion method, and experimental results show that this method has better robustness and accuracy in estimating battery SOC compared to traditional methods.
INTERNATIONAL JOURNAL OF ENERGY RESEARCH
(2022)
Article
Statistics & Probability
Mike Tsionas
Summary: This paper discusses the application of Bayesian inference in Least Median of Squares and Least Trimmed Squares, two robust techniques for handling outliers. The new Bayesian techniques are applied to linear models with independent or AR and ARMA errors. Model comparison is conducted using posterior model probabilities, and the effectiveness of the new techniques is evaluated through Monte Carlo experiments and an application to asset returns portfolios.
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
(2023)
Article
Computer Science, Theory & Methods
Gleb Beliakov, Jian-Zhang Wu
Summary: This study discusses the role of fuzzy measures in aggregation problems and the challenges of obtaining fuzzy measure coefficients from domain experts or empirical data. It introduces the concepts of k-additivity and k-maxitivity to simplify fuzzy measures, and explores the importance of learning fuzzy measures from data.
FUZZY SETS AND SYSTEMS
(2021)
Article
Computer Science, Theory & Methods
Gleb Beliakov, Francisco Javier Cabrerizo, Enrique Herrera-Viedma, Jian-Zhang Wu
Summary: The theory of capacities provides a powerful formal methodology for addressing criteria dependencies in multiple criteria decision problems. This study focuses on randomly generating capacities for simulation studies and for capacity learning through evolutionary algorithms. The results are supported by extensive numerical evidence and offer a useful tool for large scale simulations.
FUZZY SETS AND SYSTEMS
(2022)
Article
Computer Science, Theory & Methods
Tim Wilkin, Gleb Beliakov
Summary: This article discusses techniques for calculating the mode of compositional data and the challenges it faces in real-world applications, such as computational complexity and oversmoothing.
FUZZY SETS AND SYSTEMS
(2022)
Article
Computer Science, Theory & Methods
Gleb Beliakov, Marek Gagolewski, Simon James
Summary: The use of the Choquet integral allows for effective modeling of interactions and dependencies between data features or criteria in data fusion processes. This paper studies hierarchical aggregation processes that are structurally similar to feed-forward neural networks, simplifying the fitting problem and proposing a fuzzy measure method based on the Mobius representation.
FUZZY SETS AND SYSTEMS
(2022)
Correction
Computer Science, Theory & Methods
Gleb Beliakov, Enrique de Amo, Juan Fernandez-Sanchez, Manuel Ubeda-Flores
Summary: This article corrects the formula for the best-possible upper bound on the set of copulas with a given value of the Spearman's footrule coefficient that was recently published in [1].
FUZZY SETS AND SYSTEMS
(2022)
Article
Computer Science, Information Systems
Gleb Beliakov
Summary: This paper formalizes operations on capacities in matrix algebra framework, expressing various quantities characterizing input importance and dependencies through capacity derivatives. New formulas for Shapley values and nonmodularity indices are found, and relationships between Shapley interaction indices and lower order derivatives at the top and bottom elements of power sets are discovered.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Theory & Methods
Gleb Beliakov, Dmitriy Divakov
Summary: This paper outlines recent trends in capacity-based aggregation in large universes, discussing the importance of fuzzy measures in modeling input dependencies and exploring the challenges and approaches to reducing the complexity of aggregation.
FUZZY SETS AND SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Jian-Zhang Wu, Gleb Beliakov
Summary: This paper investigates the concept of nonmodularity index and its specific cases, such as Shapely and Banzhaf nonmodularity indices, which can be used to describe the interaction situations of decision criteria. The connections and differences among three categories of interaction indices are also examined. It is found that nonmodularity indices involve fewer subsets and can be more helpful in describing interaction phenomena in decision analysis.
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Gleb Beliakov, Simon James
Summary: This paper discusses the importance of measures of diversity, spread and inequality in various fields, introduces the P-D principle and ordered weighted averaging operators, and proposes the Choquet integral as a potential method for defining welfare measures. The concept of buoyancy is extended to fuzzy measures, and the optimization problem of the Choquet integral under linear constraints is explored, providing an efficient linear programming solution for antibuoyant fuzzy measures.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Computer Science, Theory & Methods
Gleb Beliakov, Enrique de Amo, Juan Fernandez-Sanchez, Manuel Ubeda-Flores
Summary: This paper investigates the pointwise best-possible bounds on the set of copulas with a given value of the Spearman's footrule coefficient. It is found that the lower bound is always a copula but the upper bound can be a copula or a proper quasi-copula, with both cases being characterized.
FUZZY SETS AND SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Gleb Beliakov
Summary: This article addresses the challenging problem of random sampling from discrete fuzzy measures. It converts the problem to sampling from an order polytope and efficiently handles it using the Markov chain random walk.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Computer Science, Information Systems
Gleb Beliakov, Jian-Zhang Wu
Summary: Capacities and the Choquet integral are powerful tools for representing decision problems with dependencies and aggregate correlated decision criteria. Random generation of suitable capacities is a vital and challenging task in this decision model. Various approaches are presented in the paper for constructing the most credible ranking of alternatives.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Gleb Beliakov, Simon James
Summary: This work focuses on solving optimization problems using the Choquet integral as the objective function, which allows for interaction between coalitions of decision variables. Efficient solution approaches are proposed for problems with a large number of variables, leveraging the antibuoyancy property and extending it to general fuzzy measures. Theoretical results are supported by numerical experiments, showing significant performance gains and scalability to a higher number of variables.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Gleb Beliakov, Thang Cao, Vicky Mak-Hau
Summary: This study addresses the problems of mutual dependence in multiple criteria decision-making and multiobjective optimization in the context of land combat vehicle selection for Australian Defence. The criteria dependencies are modeled and formalized using fuzzy measure theory. The main challenge lies in the large number of parameters quantifying all criteria interactions. Strategies such as simplifying model construction, eliciting preferences from numerical simulations and decision makers, and translating preferences into a fuzzy measure learning problem are developed. A mathematical programming problem is formulated and tested to determine fuzzy measure parameters for a relatively large number of decision criteria.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Automation & Control Systems
Juan Zhao, Tianrui Zong, Yong Xiang, Guang Hua, Xinyu Lei, Longxiang Gao, Gleb Beliakov
Summary: FSMP is a frequency spectrum modification process proposed to defend against collusion attacks, significantly degrading the perceptual quality of colluded files and deterring attackers. The method is orthogonal to traditional anti-collusion methods and can provide double-layer protection.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)