Iranian journal of fuzzy systems
Volume:20 Issue: 6, Nov -Dec 2023

The imprecision related to measurements can be managed in terms of fuzzy features, which are characterized by two components: center and spread. Outliers affect the outcome of the clustering models. In trying to overcome this problem, this paper proposes a fuzzy clustering model for L-R fuzzy data, which is based on a dissimilarity measure between each pair of fuzzy data defined as an adaptive weighted sum of the L1-norms of the centers and the spreads. The proposed method is robust based on the metric and weighting approaches. It estimates the weight of a given fuzzy feature on a given fuzzy cluster by considering the relevance of that feature to the cluster; if outlier fuzzy features are present in the dataset, it tends to assign them weights close to 0.To deeply investigate the capability of our model, i.e., alleviating undesirable effects of outlier fuzzy data, we provide a wide simulation study. We consider the ability to classify correctly and the ability to recover the true prototypes, both in the presence of outliers. The comparison made with other existing robust methods indicates that the proposed methodology is more robust to the presence of outliers than other methods. Moreover, the performance of our method decreases more slowly than others when the percentage of outliers increases. An application of the suggested method to a real-world categorical dataset is also provided.

Visual cryptography is a method used to secure images by converting them into several shares that are finally stacked to recover the original image without any calculations. Most existing techniques that can encrypt gray and color images often convert them to binary or support only limited colors, which results in reducing the quality of recovered images. Pixel expansion is another problem with existing methods. Thus, a new approach is required to encrypt gray and color images with real value, without converting them to binary or limited-color images, and also without imposing any pixel expansions. Besides, generated shares should have security, and the recovered images should be of high quality. In this research, fuzzy random grids and a meta-heuristic algorithm are used for the share generation in the encryption step. Next, the decryption step uses the fuzzy OR operator to recover high-quality images. The evaluation results demonstrate the ability of the proposed solution in encrypting gray and color images without converting them to binary, and also without pixel expansion. Besides, the results show that the proposed method is secure as individual shares do not show any information from the original image. The quality of the decrypted images has also been evaluated using subjective and objective evaluation metrics, which prove the high quality of recovered images.

In this paper we provide new several Jackson-type approximations results for continuous fuzzy-number-valued functions which improve several previous ones. We use alternative techniques adapted from Interval Analysis which rely on the gH-difference (which might not exist) and the generalized difference (which might lack the cancellation law ) of fuzzy numbers.

In a cross-docking system, doors can be used exclusively for receiving or sending operations, or they can be flexible enough to be used for both. Flexible doors have recently attracted a lot of attention as a way to improve cross-docking systems' performance. Despite the advantages of flexible doors, applying them is usually associated with additional costs due to the need for dual-function equipment to receive and send. Moreover, the distance that goods move within cross-docking facilities, from receiving to sending dock doors, significantly affects how well these facilities work. This paper presents a new bi-objective mixed-integer linear programming model for scheduling the inbound and outbound trucks in a cross-docking facility. The model objectives are to minimize total system costs, including additional costs of applying flexible doors and transshipment (maximize efficiency) and to minimize outbound trucks' tardiness from predetermined due dates (maximize responsiveness). As a result of the uncertainties in cross-dock truck scheduling problems, the parameters are considered triangular interval-valued fuzzy (IVF) numbers. Moreover, a new IVF-uncertain solution approach based on max-min operator and compromise programming concepts is proposed to solve multi-objective mathematical programming problems with triangular IVF numbers. The proposed model and IVF-solution approach are used for cross-docking operations planning at a well-known food manufacturing group. The results are also analyzed according to their sensitivity to changes in key parameters of the cross-docking problem. The comparison of the proposed approach and some fuzzy known methods for solving multi-objective models demonstrates the superior performance of the proposed approach in the real case study.

We introduce the notions of Sasaki mapping and of sharp elements on a d0-algebra. We investigate the relationships of sharp elements with Sasaki mappings and with central elements, thus generalizing some results known for D-lattices. Namely, we give a characterization of sharp elements, by means of Sasaki mappings, which extends a result of Bennett and Foulis; we also prove that an element is central if and only if it is a sharp element and every element is compatible with it: this generalizes a result of Riecanová.

In this paper, our aim is to make a novel interpretation of relation between $(\overline{N},p,q)$ method and $P$-convergence for double sequences of fuzzy numbers. In accordance with this aim, we derive some Tauberian conditions, controlling $O$-oscillatory behavior of a double sequence of fuzzy numbers in certain senses, from $(\overline{N},p,q)$ summability to $P$- convergence with some restrictions on the weight sequences. As special cases, we indicate that $O$-condition of Hardy type with respect to $(P_m)$ and $(Q_n)$ are Tauberian conditions for $(\overline{N},p,q)$ summability under some additional conditions. In the sequel, we prove a fuzzy Korovkin-type approximation theorem by using the $(\overline{N},p,q)$ summability method for fuzzy positive linear operators.

Given a uniform space (X, U), we introduce, from the uniformity U, some uniformities on the set F(X) of all normal, upper semicontinuous with compact support fuzzy sets on X: the Skorokhod uniformity, the level-wise uniformity, the endograph uniformity and the sendograph uniformity. For metric spaces we prove that these uniformities coincide with the uniformities induced by the Skorokhod metric (the level-wise metric, the endograph metric and the sendograph metric, respectively). We study completeness of this class of uniform spaces.

Ranking fuzzy numbers have become of growing importance in recent years, especially as decision-making is increasingly performed under greater uncertainty. In this paper, we extend the concept of magnitude to rank fuzzy numbers to a more general definition to increase in flexibility and generality. More precisely, we propose a multidimensional approach to rank fuzzy numbers considering alternative magnitude definitions with three novel features: multidimensionality, normalization, and a ranking based on a parametric distance function. A multidimensional magnitude definition allows us to consider multiple attributes to represent and rank fuzzy numbers. Normalization prevents meaningless comparison among attributes due to scaling problems, and the use of the parametric Minkowski distance function becomes a more general and flexible ranking approach. The main contribution of our multidimensional approach is the representation of a fuzzy number as a point in a $n$-dimensional normalized space of attributes in which the distance to the origin is the magnitude value. We illustrate our methodology and provide further insights into different normalization approaches and parameters through several numerical examples. Finally, we describe an application of our ranking approach to a multicriteria decision-making problem within an economic context in which the main goal is to rank a set of credit applicants considering different financial ratios used as evaluation criteria.

In data mining, classification is one of the most important steps in predicting the target class. Classification is performed by an improved model in existing work in which feature selection is performed based on the bat optimization method to increase the classification accuracy. And an Enhanced Neural Network is used for classification which includes Intuitive, Interpretable Correlated-Contours fuzzy rules. And an effective model is created based on the extraction of fuzzy rules, where data partitioning is performed via a similarity-based directional component. However, the dataset used for experimentation is noisy as well as incomplete data values. Due to incompleteness, knowledge discovery is obstructed and the result of classification is affected as well. And bat provides very slow convergence and easily falls into local optima. To solve this issue, an improved framework is introduced in which missing value imputation is performed by using k means clustering, and then for feature selection, an improved cuckoo search optimization is used. An enhanced classifier based on fuzzy logic and Alex Net neural network structure (F-ANNS) is used for classification and hybrid Ant Colony Particle Swarm Optimization (HASO) is used for optimizing parameters of the AlexNet neural network classifier. The results show that the proposed work is more effective in precision, recall, accuracy, and f-measure as shown by experimental results.

This paper introduces a novel adaptive nonsingular fast terminal sliding mode approach that benefits from an interval type-2 fuzzy logic estimator and a gain for control and synchronization of chaotic systems in the presence of uncertainty. The nonsingular fast terminal sliding mode controller is developed to increase the convergence rate and remove the singularity problem of the system. Using the proposed method, the finite-time convergence has been ensured. To eliminate the chattering phenomenon in the conventional sliding mode controller, the discontinuous sign function is estimated using an interval type-2 fuzzy inference system (FIS) based on the center of sets type reduction followed by defuzzification. By adding the proportionate gain to the interval type-2 FIS, the robustness and speed of the controller system is enhanced. An appropriate Lyapunov function is utilized to ensure the closed-loop stability of the control system. The performance of the controller is evaluated for a nonlinear time-varying second-order magnetic space-craft chaotic system with different initial conditions in the presence of uncertainty. The simulation results show the efficacy of the proposed approach for the tracking control problems. The time and frequency domain analysis of the control signal demonstrates that the chattering phenomenon is successfully diminished.

Network congestion is one of the major issues in wireless sensor networks (WSNs) that result in packet loss, reduced network lifetime, low throughput and energy waste. Determining a better path to mitigate the congestion is a better approach to improve the performance of WSNs. In this paper, an adaptive neuro-fuzzy inference system (ANFIS) based path determination approach is proposed to mitigate the congestion with black widow optimization (BWO) algorithm. The proposed approach first develops a framework to mitigate the congestion in WSNs. Then it forecast the buffer occupancy with the exponential smoothing technique. Finally, ANFIS is applied in the proposed approach for determining the path with appropriate weights by considering the remaining energy, hop count and buffer occupancy. Here, the hop count, buffer occupancy and remaining energy are considered as the input factors for the ANFIS. The simulation results of the proposed method show better quality of service, high energy, low delay, high packet delivery ratio with number of increasing alive nodes when compared to existing methods.