Marzieh Rezaei; Abbas Parchami; Ayyub Sheikhi
Articles in Press, Accepted Manuscript, Available Online from 23 May 2026
Abstract
Clustering is one of the data mining tools frequently used in multivariate methods to group observations into clusters based on the similarity of variables. While many clustering algorithms have been developed for data extracted from well-defined (crisp) populations, numerous ...
Read More
Clustering is one of the data mining tools frequently used in multivariate methods to group observations into clusters based on the similarity of variables. While many clustering algorithms have been developed for data extracted from well-defined (crisp) populations, numerous real-world problems involve inherently fuzzy populations such as high-consumption households or high-quality products. In such cases, each data point is described not only by its observed values but also by its degree of membership to the fuzzy population. This study proposes a novel clustering algorithm specifically designed for data extracted from fuzzy populations. The proposed method generalizes the classical k-means algorithm by considering membership degrees in the clustering process. The effectiveness of the proposed algorithm was evaluated on both synthetic and real-world datasets using several validity indices. According to these indices, the algorithm demonstrates satisfactory clustering quality and computational efficiency. The analysis of the proposed algorithm in the presence of outliers shows that, although outliers can slightly shift cluster centers, the overall clustering structure remains stable when clusters are well-separated.
M. S. Abolhassanifar; R. Saadati; B. Pazar Varol
Articles in Press, Accepted Manuscript, Available Online from 24 September 2026
Abstract
In this paper, we study pseudo-Fredholm equations in generator-based ∗–fuzzy measure spaces. A control radius is defined from the generator and the ∗-fuzzy integral, and a Boyd-Wong-type fixed-point theorem is proved. We obtain residual, inexact-iteration, and data-stability ...
Read More
In this paper, we study pseudo-Fredholm equations in generator-based ∗–fuzzy measure spaces. A control radius is defined from the generator and the ∗-fuzzy integral, and a Boyd-Wong-type fixed-point theorem is proved. We obtain residual, inexact-iteration, and data-stability estimates. For the Hamacher triangular norm, the pseudo-Nyström scheme is second- order convergent. Changes of generator, perturbations of the ∗-fuzzy measure, and a bi-scale L1–L2 generator error construction are also considered.
Ghulam Muhammad; Qingyu Zhang; Muhammad Akram; Tofigh Allahviranloo
Articles in Press, Accepted Manuscript, Available Online from 28 September 2026
Abstract
This article develops a mathematical framework for fuzzy fractional differential system using the concept of granular Caputo fractional derivative with respect to another function ψ within the horizontal membership function (HMF)-based granular differentiability framework. The propose ...
Read More
This article develops a mathematical framework for fuzzy fractional differential system using the concept of granular Caputo fractional derivative with respect to another function ψ within the horizontal membership function (HMF)-based granular differentiability framework. The propose technique addresses the need for a unified approach that simultaneously incorporate the function ψ as memory kernels for flexible modeling, multi-dimensional representation of uncertainty through HMFs, and analytical fuzzy solution using granular fuzzy Laplace transform with respect to another function (LTWR-AF). First, we introduce the granular ψ-fuzzy fractional integral (ψ-FFI), granular ψ-Caputo fractional derivative (ψ-CFD) and granular fuzzy LTWR-AF. Fundamental properties, results, and relationships among these properties are established in HMF representation. Based on this foundation, closed-form analytical fuzzy solution (AFS) of granular Caputo fuzzy fractional differential equations (GCFFDEs) involving granular ψ-CFD are determined and expressed compactly in terms of Mittag-Leffler function (M-LF). Moreover, the existence and uniqueness (E&U) of the fuzzy solutions for the abstract granular Caputo fuzzy fractional differential system (GCFFDS) with fuzzy parameters and initial data are examined under mild conditions. To illustrate the applicability of the proposed framework, the proposed methodology is applied to the granular Caputo fuzzy fractional sustainable, impact-intensive, quarantined, and recycled/remanufactured (SIQR) supply-chain model under environmental disruptions. Numerical simulations and graphical representations demonstrate that the proposed model effectively describe the uncertainty propagation and memory-dependent behavior in supply-chain dynamics.
Van Binh Ngo; Loc Tien Le
Articles in Press, Accepted Manuscript, Available Online from 28 September 2026
Abstract
Car-following systems play a critical role in advanced driver assistance systems and autonomous driving applications, where accurate detection and stable tracking of the preceding vehicle are essential for ensuring safety and smooth motion. This paper presents a low-cost vision-based car-following ...
Read More
Car-following systems play a critical role in advanced driver assistance systems and autonomous driving applications, where accurate detection and stable tracking of the preceding vehicle are essential for ensuring safety and smooth motion. This paper presents a low-cost vision-based car-following control system that combines monocular vision sensing with an intelligent speed controller. A convolutional neural network is employed to detect the target vehicle and extract spatial information from input images, while the measured distance is refined using a Kalman filter to suppress noise and improve estimation stability. Based on the visual feedback, the vehicle speed is regulated by a Recurrent Multilayer Interval Type-2 Fuzzy Neural Network (RMIT2FNN), enabling adaptive responses to the motion of the leading vehicle and maintenance of a safe following distance. The proposed approach is first validated through simulations, in which its tracking performance is compared with conventional FNN and IT2FNN controllers under the same operating conditions. The simulation results demonstrate improved tracking accuracy and rapid responses to velocity variations. Subsequently, real-time experimental validation is conducted on a 1/10-scale vehicle platform, confirming that the system can detect and track the preceding vehicle stably while maintaining the desired following distance under indoor operating conditions. These findings provide preliminary evidence of the feasibility of the proposed method as a cost- efficient vision-based car-following solution under controlled experimental conditions.
Jyun-Yu Jhang; Cheng-Jian Lin; Xue-Qian Lin; Han Cheng
Articles in Press, Accepted Manuscript, Available Online from 01 October 2026
Abstract
As deep learning applications continue to expand, the performance of network models becomes increasingly critical. To enhance model accuracy, training often relies on a large amount of labeled data. However, manually labeling data is both labor-intensive and expensive. In recent years, Domain ...
Read More
As deep learning applications continue to expand, the performance of network models becomes increasingly critical. To enhance model accuracy, training often relies on a large amount of labeled data. However, manually labeling data is both labor-intensive and expensive. In recent years, Domain Adaptation (DA) techniques have gained significant attention as they enable models to be trained with limited labeled data while maintaining performance on unlabeled data. In particular, Unsupervised Domain Adaptation (UDA) faces a major challenge due to differences in data distributions. Addressing the problem of domain shift and identifying domain-invariant features with limited information remains a crucial task. This paper proposes a fuzzy-theory-based unsupervised domain adaptation approach called the Deep Fuzzy- Similarity-based Local Domain Adaptation Network (DFS-LDAN). The proposed method aims to mitigate domain shift and reduce annotation costs. DFS-LDAN integrates fuzzy similarity and Maximum Mean Discrepancy (MMD), allowing the model to retain more soft information during learning. This enhances its flexibility in handling uncertainties in target domain data. Additionally, this study introduces a Local Domain Adaptation strategy that aligns distributions at the class level. It employs Fuzzy C-Means (FCM) clustering to adaptively adjust feature weights based on membership degrees, further improving the model’s generalization capability. The effectiveness of DFS-LDAN is evaluated on three public datasets: Office-31, ImageCLEF-DA, and Office-Home. Comparative experiments with various domain adaptation methods demonstrate the superiority of DFS-LDAN. The results show that the proposed method, across the best-performing variant, achieves average accuracies of 89.1% (via Product-Sum), 90.4% (via Product-Sum), and 68.4% (via Min-Max), respectively.
J. Tayyebi; E. Hosseinzade; M. Niksirat
Articles in Press, Accepted Manuscript, Available Online from 17 October 2026
Abstract
The assignment problem is a foundational combinatorial optimization problem that requires a decision maker to allocate agents to tasks in order to minimize the total cost incurred. From a graph-theoretic perspective, this problem can be framed as a matching problem on a bipartite graph. This paper presents ...
Read More
The assignment problem is a foundational combinatorial optimization problem that requires a decision maker to allocate agents to tasks in order to minimize the total cost incurred. From a graph-theoretic perspective, this problem can be framed as a matching problem on a bipartite graph. This paper presents a natural extension of the traditional assignment problem by considering fuzzy graphs as the underlying framework. This extension proves significant in practical situations where expert opinions contribute to assessing the relative significance level of agents (or tasks) and the satisfaction levels associated with different assignments. The main goal for the decision maker is not just to minimize costs but also to consider expert opinions. This paper addresses this problem by first formulating it as a tri-objective optimization problem. Subsequently, the ϵ-constraint method is applied to propose an interactive algorithm designed to identify satisfactory solutions. Finally, a practical example of assigning teachers to schools is provided to demonstrate the algorithm in a real-world application.
J. Zheng; Y. M. Wang; K. Zhang; J. Q. Gao
Articles in Press, Accepted Manuscript, Available Online from 26 October 2026
Abstract
Case-driven fuzzy reasoning (CDFR) enhances emergency response efficiency by leveraging historical experience and evaluation information. However, existing CDFR models lack a systematic framework for multi-source information fusion and fail to adequately address the fuzziness regarding information ...
Read More
Case-driven fuzzy reasoning (CDFR) enhances emergency response efficiency by leveraging historical experience and evaluation information. However, existing CDFR models lack a systematic framework for multi-source information fusion and fail to adequately address the fuzziness regarding information reliability and inter-group consensus. To bridge this gap, this paper proposes a CDFR framework for emergency decision-making under interval probabilistic linguistic information. The reasoning process commences with fuzzy case retrieval to establish the decision context. Multi-source information is then pre-processed through reliability measure and consensus-driven adjustment mechanism to ensure the credibility and consistency of the evaluation information. Furthermore, attribute weights are derived via the IPLTS- embedded DEMATEL method, and group-wise evaluation information is then fused using the IPLTSWG operator. Finally, an interval-based GLDS approach is proposed to generate the final ranking, which preserves the uncertainty of the original information throughout the reasoning chain. The practicability of the proposed framework is demonstrated through a case study of public health emergency response.
H. Ghasemi; A. Akbarzadeh Kalat
Articles in Press, Accepted Manuscript, Available Online from 28 October 2026
Abstract
This paper develops a robust indirect adaptive fuzzy output-feedback control scheme for attitude tracking of a rigid satellite in the presence of unknown inertia matrix, completely unknown nonlinear dynamics, and external disturbances. Unlike state-feedback approaches, the proposed method ...
Read More
This paper develops a robust indirect adaptive fuzzy output-feedback control scheme for attitude tracking of a rigid satellite in the presence of unknown inertia matrix, completely unknown nonlinear dynamics, and external disturbances. Unlike state-feedback approaches, the proposed method uses only measurable attitude outputs and does not require direct angular-velocity measurements. First, the Modified Rodrigues Parameters (MRPs) model of satellite dynamic is reformulated into a new state-space representation suitable for the integrated design of the adaptive fuzzy observer and controller. A robust adaptive fuzzy observer is developed to estimate the unavailable state variables from the measured outputs and approximate the unknown nonlinear dynamics together with the unknown inertia matrix in the presence of uncertainties and bounded disturbances. Based on the estimated information, an adaptive fuzzy output-feedback controller is designed to track a desired attitude trajectory. Lyapunov-based design establish the stability of the proposed observer and closed-loop control system. Numerical simulations demonstrate accurate attitude tracking, effective state and uncertainty estimation, and robustness against external disturbances. The proposed framework therefore reduces dependence on angular-velocity sensors and prior knowledge of the satellite dynamics while maintaining reliable tracking performance.
)
