International Journal of Fuzzy Systems最新文献

In the era of big data, there exist complex structure between different classes labels. Hierarchical structure, among others, has become a representative one, which is mathematically depicted as a tree-like structure or directed acyclic graph. Most studies in the literature focus on static feature selection in hierarchical information system. In this study, in order to solve the incremental feature selection problem of hierarchical classification in a dynamic environment, we develop two incremental algorithms for this purpose (IHFSGR-1 and IHFSGR-2 for short). As a preliminary step, we propose a new uncertainty measure to quantify the amount of information contained in the hierarchical classification system, and based on this, we develop a non-incremental hierarchical feature selection algorithm. Next, we investigate the updating mechanism of this uncertainty measure upon the arrival of samples, and propose two strategies for adding and deleting features, leading to the development of two incremental algorithms. Finally, we conduct some comparative experiments with several non-incremental algorithms. The experimental results suggest that compared with several non-incremental algorithms, our incremental algorithms can achieve better performance in terms of the classification accuracy and two hierarchical evaluation metrics, and can significantly accelerate the fuzzy rough set-based hierarchical feature selection.

The role of transportation in international trade cannot be overlooked; hence, the need for regular upgrades of roads and marketplaces. The Karakorum Highway (KKH), a vital part of the China–Pakistan economic corridor that connects China with Arabian waters, has not received significant attention from the National Highway Authority (NHA) due to uncertainties in many regions. To address this issue, the study employs Dombi operations using Polytopic fuzzy sets to explore uncertainty in decision-making. The Dombi t-norm and t-co-norm can capture inconsistencies, making it an effective tool in the decision-making process. The study applies the Polytopic fuzzy Dombi-weighted averaging (PF-DWA) operator, the Polytopic fuzzy Dombi-ordered weighted averaging (PF-DOWA) operator, and the Polytopic fuzzy Dombi hybrid-weighted averaging (PF-DHWA) operator to demonstrate how the model can help the NHA open bidding for interested companies to repair damaged areas, bridges, and side barriers affected by floods. The study reveals that infrastructure is essential for the development of any country, and the most suitable choice for reconstruction can be made using the proposed methods.

In this article, an adaptive prescribed performance tracking control scheme is proposed for switched nonlinear systems with output dead zone and unmeasured state variables using an adaptive fuzzy approach. Fuzzy logic systems are utilized to learn the unknown nonlinear functions. The output nonlinearity is resolved via introducing Nussbaum function. The novelty of this article is that a shift function is utilized to break the strict restriction that the initial value of the tracking error must be within the initial value of the finite-time performance function. In addition, a switched observer is adopted to reduce the conservativeness caused by the use of a common observer. Then, by combining the average dwell time scheme and the backstepping technology, a novel observer-based fuzzy adaptive controller is developed, which can assure that all the closed-loop signals of the switched systems are bounded under a type of slowly switching signals and the tracking error converges to a pre-specified range in finite time even if the initial value of the tracking error is greater than the performance function. Finally, the simulation results are shown to verify the feasibility of the presented control scheme.

The additive ratio assessment system (ARAS) method is an effective technique for simplifying complex decision problems by determining the optimal alternative through the relative index (utility degree) to the ideal solution. However, there are still some shortcomings in the existing researches on the extension of this method when it is utilized in different decision environments, such as ignoring the correlation relationship between attributes, the lack of flexibility in the utilization of the decision process, and the relative index to the ideal solution may be scaled up or down with the ratio form. In order to overcome these disadvantages, this paper proposes the novel T-spherical fuzzy (TSF) cross entropy (TSFCE) measure and T-spherical Aczel-Alsina Heronian mean (TSFAAHM) aggregation operators and uses them to improve the ARAS method in the TSF environment. For the TSF multiple attribute group decision-making (MAGDM) problems, a group decision making model based on the improved ARAS is designed. In this model, the experts’ weights are obtained by the TSFCE-based similarity measure. The attribute combined weights are calculated by fusing the objective weights obtained by TSFCE-based entropy measure and the subjective weights got by the extended stepwise weight assessment ratio analysis (SWARA) integrated with TSFCE. In the improved ARAS method, the T-spherical Aczel-Alsina Weighted Heronian mean (TSFAAWHM) operator can capture the correlation relationship between the attributes. Compared with the relative index, the TSFCE can reflect the difference between the alternatives and the ideal solution to obtain a more stable solution ranking. Lastly, an illustrative example about the sustainable supplier selection of power battery echelon utilization (PBEU) for a 5G base station is used to demonstrate the proposed method. The effectiveness, practicability and superiority of proposed method are illustrated by parameters influence and methods comparison analysis.

In this work, the problem of sliding mode tracking control with previewable reference signals is investigated for nonlinear discrete-time T–S fuzzy multi-agent systems (MASs) with time-delays. First, an augmented error system (AES) including the error as well as the previewable reference signal is constructed by using the theory of preview control. Subsequently, a fuzzy sliding mode surface (SMS) is proposed for the AES. Next, a general scheme for stability analysis is given for the sliding motion of the considered system. A sliding mode controller with preview action satisfying the discrete-time reachability condition is designed. At the end, the cases of arithmetic are offered to demonstrate that the suggested control strategy can effectively enhance the tracking performance of the MAS’s output with respect to the reference signal.

Formal Concept Analysis (FCA) is an order theory-based methodology employed for concept analysis and construction. Incomplete fuzzy formal context is employed to present the uncertainty or lack of memberships between individuals and attributes. Acceptable implications and necessary implications are two types of implications that assess the validity of knowledge within incomplete formal contexts. On the one hand, attribute exploration approaches within incomplete formal contexts rely on the prior knowledge of experts. On the other hand, in the existing reasoning mechanism for acceptable implications and necessary implications, the bases are inconvenient as they recursively involve the bases of all the completions of the incomplete formal context. Another critical issue is that the inference rules, originally apply to the implications in formal contexts, may yield invalid implications when they are applied to the two types of implications. In this paper, we firstly discretize incomplete fuzzy formal context into incomplete formal context by employing a dual-threshold filter function and then model the incomplete formal context by two specially constructed decision contexts. Next, we re-represent acceptable implications and necessary implications based on decision implications and demonstrate that the inference rules Augmentation and Combination, initially designed for decision implications, are practicable for necessary implications and acceptable implications. Furthermore, we utilize Augmentation, Combination, and another inference rule Reflexivity to jointly define the completeness and non-redundancy for sets of necessary implications and that of acceptable implications. Finally, we establish necessary implication basis and acceptable implication basis, which preserve all the information implied in the two types of implications while simultaneously minimizing the total number of implications.

This paper introduces a novel control method for the quadrotor suspension system, addressing the challenges posed by a suspended load, external disturbances, input saturation and model dynamic uncertainty. The primary goal of this method is to achieve precise quadrotor trajectory tracking while minimizing oscillations in the suspended load. To model the system, the Udwadia-Kalaba equation is employed to handle the interaction between the quadrotor and the suspended load. Input saturation is mitigated using the hyperbolic tangent function, and a trapezoidal acceleration algorithm is utilized for trajectory design. To deal with composite disturbances, an adaptive fuzzy control method is developed and a double closed-loop nonlinear control method ensures system stability based on the Lyapunov stabilization criterion. Simulation results confirm the method’s effectiveness in accurately regulating the quadrotor system in the presence of external disturbances and model uncertainties, while also reducing suspended load oscillations under input saturation conditions.

Trust network analysis has been widely applied in various fields, such as group recommendation, group decision-making and other related areas. In this paper, we focus on obtaining the complete trust network in which experts express their trust relationships for another with a single linguistic term or distribution assessments of a linguistic term set. We first discuss the conditions of obtaining the complete trust network, and the propagation and aggregation of the trust relationships with a single linguistic term. Since the linguistic term set may be symmetric and uniform, symmetric and non-uniform, or asymmetric and non-uniform, we translate linguistic terms into numerical indexes and define the propagation operator based on the semantics of the linguistic term and the Archimedean t-norm. The propagation result is translated to 2−tuple linguistic model because it may not exist in the initial linguistic term set. Some properties are proposed to verify that the proposed operator is compatible with human thought. Then the 2−tuple distribution assessments on a linguistic term set are defined, and the other aggregation operator is proposed to propagate linguistic distribution assessment trust relationships. The second aggregation operator focuses on both the aggregation of linguistic terms and symbolic proportions of linguistic terms and is a generalization of the first operator. Finally, a numerical example of CouchSurfing comparative analyses further demonstrates that the proposed operators are effective and reasonable, and can consider the different semantics of a linguistic term in practical application.

This study is concerned with a realization of horizontal federated Gustafson–Kessel clustering algorithm and the ensuing construction of ellipsoidal information granules. As a fundamental component of Granular Computing, information granules play an important role in human-centric computing, such as human cognition and decision-making. Driven by the concerns of data privacy and confidentiality, it is of interest to investigate how to construct information granules on the basis of horizontally partitioned numeric data distributed across different sites using a privacy-preserving approach. To meet this challenge, federated learning has become an appealing solution to the problem of forming meaningful clusters (information granules) while ensuring data privacy and confidentiality. A two-development strategy is applied in the proposed algorithm: first, a collection of numeric representatives (prototypes) is obtained with the use of federated Gustafson–Kessel algorithm, which is able to reveal ellipsoidal shapes in the datasets and second, information granules are built through engaging the principle of justifiable granularity. A series of experimental studies demonstrate the effectiveness of the proposed federated Gustafson-Kessel algorithm in revealing the structure of the entire dataset. The formed ellipsoidal information granules help us gain a better insight into the topology of the overall dataset.

In practical decision-making, linguistic term set is a useful tool to describe the uncertainty and fuzziness of data sources. However, in some decisions, when the data source is unreliable or the decision involves future factors, the evaluation given by the linguistic term set will have a certain degree of error. This paper proposes a binary risk linguistic set based on linguistic term set and R-set. The binary risk linguistic set considers the linguistic term set and the risk factors that may lead to errors in language evaluation. In order to facilitate the use of binary risk linguistic set, the risk conversion function and operational laws are introduced. Next, since group decision-making involves multiple experts, considering the social relations between experts, a method to estimate the missing values in the social network matrix is proposed by utilizing the trust intensity propagation operator and the relationship intensity propagation operator. Risk perception can reflect the subjective judgment of experts on the characteristics and severity of a particular risk, and different judgment results can reflect the attitude of experts to risk. Hereby, this study proposes a risk clustering method based on the risk perception of experts. Furthermore, we propose an adaptive weight updating method based on social network matrix. Then, a binary risk linguistic fuzzy behavioral TOPSIS method is proposed to deal with the multi-attribute large-scale group decision-making (MALSGDM) problem. Finally, a case study is used to demonstrate the feasibility of the presented method, and its effectiveness is validated through comparison with other MALSGDM methods. To demonstrate the effectiveness of the proposed method, this study also perform sensitivity and stability assessments of the decision-makers’ weight and behavior characteristics.