Article
Computer Science, Artificial Intelligence
Yaser Donyatalab, Fatma Kutlu Gundog, Fariba Farid, Seyed Amin Seyfi-Shishavan, Elmira Farrokhizadeh, Cengiz Kahraman
Summary: Spherical fuzzy sets have gained popularity in various fields as a generalization of picture fuzzy sets and Pythagorean fuzzy sets. This study proposes novel distances and similarity measures for spherical fuzzy sets and applies them to medical diagnosis for COVID-19. The newly defined similarity measures provide advantages and contribute to the understanding of similarity between objects.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Onesfole Kurama
Summary: This study extends the similarity classifier to include Dombi aggregation operators, achieving improved classification accuracies on medical datasets compared to earlier methods. The proposed method shows significant improvements on fertility, lung cancer, Haberman's survival, and liver disorder datasets.
PATTERN RECOGNITION LETTERS
(2021)
Article
Mathematics
Rodrigo Gomez Monge, Evaristo Galeana Figueroa, Victor G. Alfaro-Garcia, Jose M. Merigo, Ronald R. Yager
Summary: This paper introduces variance logarithmic averaging operators and analyzes their properties, families, and particular cases, providing an illustrative example from financial markets to showcase the design of these operators. Results show that the use of variance measures aids decision-making by offering new tools for information analysis and extends the available tools for decision-making under ignorance, uncertainty, and subjective environments.
Article
Computer Science, Artificial Intelligence
Yaser Donyatalab, Fariba Farid, Fatma Kutlu Gundogdu, Elmira Farrokhizadeh, Seyed Amin Seyfi Shishavan, Cengiz Kahraman
Summary: The newly developed three-dimensional spherical fuzzy sets are effective in handling uncertainty and quantifying expert judgments, with various applications such as medical diagnosis and pattern recognition. Novel distance and similarity measures for spherical fuzzy sets have been proposed in this study, with applications in pattern recognition for COVID-19 virus.
JOURNAL OF MULTIPLE-VALUED LOGIC AND SOFT COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Vikas Srivastava, Amar Kishor, Amit K. Singh
Summary: In this paper, a novel representative of the existing family of ordered weighted aggregation (OWA) operators with constant orness is presented. The proposed OWA operator is based on the beta function and has two different types of operators with respect to optimism and pessimism. It provides flexibility for decision making and comparisons with other OWA operators are discussed. The properties and applications of the proposed operators are also analyzed in detail.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2023)
Article
Computer Science, Artificial Intelligence
Shuvasree Karmakar, Mijanur Rahaman Seikh, Oscar Castillo
Summary: This study aims to develop matrix games using Type-2 Intuitionistic Fuzzy Sets (T2IFS) as an extension to bridge the gap in portraying decision-makers' fuzzy preferences under different parameters, considering both acceptance and non-acceptance. The proposed methodology includes defining Hamacher aggregation operators in T2IFS environment, proposing Minkowski distance of T2IFS based on the Hausdorff metric, forming a similarity measure of T2IFS, and solving matrix games using the proposed distance measure with the biogas-plant implementation problem to validate applicability and validity.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Information Systems
Marek Gagolewski, Maciej Bartoszuk, Anna Cena
Summary: Internal cluster validity measures are often used to determine the appropriate number of partitions for a dataset. However, treating these measures as objective functions in unsupervised learning activities may not always lead to meaningful results. Many validity indices do not align well with expert knowledge. Meanwhile, a new variant of the Dunn index, incorporating OWA operators and near-neighbour graphs, has shown promising performance in separating subspaces of higher density.
INFORMATION SCIENCES
(2021)
Article
Mathematics
Emili Vizuete-Luciano, Sefa Boria-Reverter, Jose M. Merigo-Lindahl, Anna Maria Gil-Lafuente, Maria Luisa Sole-Moro
Summary: This paper introduces a new assignment algorithm using the OWA operator and its extensions in the Branch-and-bound algorithm, providing more detailed information. The algorithm is applied in a consumer decision-making model in Barcelona, aiding in selecting grocery districts that best suit their needs, while considering different sources of information independently.
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
Environmental Sciences
Alex Zabeo, Gianpietro Basei, Georgia Tsiliki, Willie Peijnenburg, Danail Hristozov
Summary: The proposed novel procedure for similarity assessment and grouping of nanomaterials is based on various metrics and hierarchical clustering, ensuring robust grouping independent of the dataset used, and a software tool has been developed to facilitate its application.
Article
Mathematics
Dejan G. Ciric, Zoran H. Peric, Nikola J. Vucic, Miljan P. Miletic
Summary: The sounds of industrial products carry important information and can be used for product classification and malfunction detection. This paper maps the sounds of seven industrial products into mel-spectrograms and investigates the similarities within and between different classes. Three commonly used image similarity measures, including Euclidean distance, Pearson correlation coefficient, and structural similarity index, are compared to analyze their behaviors. The results show that five classes have similar mel-spectrograms, while two classes have unique properties with significantly larger similarity within the class compared to between classes. The applied image similarity measures lead to similar general results but show differences in the relationship of similarity among classes, especially with the use of SSIM.
Article
Computer Science, Artificial Intelligence
Eva Blanco-Mallo, Laura Moran-Fernandez, Beatriz Remeseiro, Veronica Bolon-Canedo
Summary: This paper provides an overview of the use of distance measures in machine learning and data mining tasks. It examines the factors to consider when choosing the most appropriate measure and analyzes seven commonly used measures, exploring their properties and applications. The paper also conducts experiments to study their relationships, performance, and their performance in the presence of noise, as well as the execution time required by each measure.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Weiwei Li, Pingtao Yi, Lingyu Li
Summary: This paper introduces a new aggregation operator (CB-IOWA) that allows alternatives to highlight their competitive advantage and calculates final weights to reflect this advantage, discussing its main properties and providing an application example to illustrate its usage.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Raul Perez-Fernandez, Gustavo Ochoa, Susana Montes, Irene Diaz, Javier Fernandez, Daniel Paternain, Humberto Bustince
Summary: The notion of orness measure for aggregation functions has a history dating back to the early 1970s and has since evolved to include axiomatic definitions for various types of aggregation functions, with proposed construction methods and detailed studies on specific examples like the discrete Choquet integral and uninorms.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Murat Kirisci
Summary: This article introduces the use of distance and cosine similarity metrics to measure the similarity and difference between two sets, as well as new metrics of distance and cosine similarity among Fermatean fuzzy sets. Methods for constructing other similarity measures based on cosine similarity and Euclidean distance are proposed, and a cosine distance measure is obtained. Practical examples are provided to illustrate the reasonableness and effectiveness of the proposed method.
KNOWLEDGE AND INFORMATION SYSTEMS
(2023)
