Computer Science
Mursubai Sandhya Rani; Raghavendra Sai
Volume 22, Issue 2 , March and April 2025, Pages 1-24
Abstract
The explosion of data in the last ten years has led to a substantial focus on big data (BD) in information area. The philosophical applications of "query optimization (QO)" are crucial in BD environments' data retrieval processes. Several distributed data processing platforms in cloud were developed ...
Read More
The explosion of data in the last ten years has led to a substantial focus on big data (BD) in information area. The philosophical applications of "query optimization (QO)" are crucial in BD environments' data retrieval processes. Several distributed data processing platforms in cloud were developed to provide BD query optimization services that are both affordable and effective. Nevertheless, due to a lack of consideration for energy-related concerns and query characteristics, most solutions resulted in higher "energy consumption (EC)" and lower accuracy. We introduced an innovative deep-learning approach to arrange big data to overcome the issue. This work presents an effective query optimization that uses the Quantum parallel multi-layer Monte Carlo optimization method (QPMMCOA) optimizer and a load balancer based on Bayesian fuzzy clustering to address the problems associated with query optimization process. There are two phases to the suggested technique: (1) Big data arrangement and (2) Query Optimization. The first step arranges BD using preprocessing, feature extraction, feature selection, and deep learning-based BD arrangement. The improved Deep Residual Shrinkage Network (IDRSN) algorithm is used for the BD arrangement. The essential features are selected using the Chaotic Vertex Search algorithm (CVSA). During the second phase, a Bayesian fuzzy clustering-based load balancer is used with the QPMMCOA optimizer to improve overall query processing performance and ignore energy-efficient query plans. At last, the process of evaluating similarity is carried out. The experimental results demonstrated that the method performed better than other existing algorithms.
Engineering
Loc Tien Le; Long Kim Ngo
Volume 22, Issue 2 , March and April 2025, Pages 25-39
Abstract
This paper proposes a new solution for controlling complex nonlinear systems, through the combination of a type 2 fuzzy CMAC controller and Jordan neural network. This method takes advantage of type 2 fuzzy CMAC in dealing with uncertainties and learning ability, while the Jordan neural network helps ...
Read More
This paper proposes a new solution for controlling complex nonlinear systems, through the combination of a type 2 fuzzy CMAC controller and Jordan neural network. This method takes advantage of type 2 fuzzy CMAC in dealing with uncertainties and learning ability, while the Jordan neural network helps to enhance the stability and improve the performance of the system. The adaptive learning laws were designed to help the proposed network automatically update the network parameters. The results from simulations and experiments have shown that this method achieves superior accuracy and robustness compared to other methods. When applied to control the Magnetic Levitation System, this method shows great potential in solving complex nonlinear control problems, opening up new approaches in this field.
Sogol Niazian; Mona Aaly Kologani; R. A. Borzooei
Volume 22, Issue 2 , March and April 2025, Pages 41-57
Abstract
In this paper, by using the notion of ordinal sum of equality algebras which is defined in [4], weinvestigate some related properties and different kinds of filters of it such as prime and maximal. Inaddition, by focusing on chain equality algebra, we define the concept of a cut of an equality algebraand ...
Read More
In this paper, by using the notion of ordinal sum of equality algebras which is defined in [4], weinvestigate some related properties and different kinds of filters of it such as prime and maximal. Inaddition, by focusing on chain equality algebra, we define the concept of a cut of an equality algebraand use it to introduce a new ordinal sum of them. By defining a relation relative to a cut, we showthat the family of classes is made a cone algebra. Finally, we prove that every chain equality algebracan be represented as an ordinal sum of chain cone algebras.
Fuzzy approximation
Omid Mehrabi; Ahmad Fakharian; Mehdi Siahi; Amin Ramezani
Volume 22, Issue 2 , March and April 2025, Pages 59-80
Abstract
In this paper, a new form of critic-only Reinforcement Learning algorithm for continuous state spaces control problems is proposed. Our approach, called Fuzzy-RBF Least Square Policy Iteration (FRLSPI), tunes the weight parameters of the fuzzy-RBF network (a hybrid model constituted by combining Takagi-Sugeno ...
Read More
In this paper, a new form of critic-only Reinforcement Learning algorithm for continuous state spaces control problems is proposed. Our approach, called Fuzzy-RBF Least Square Policy Iteration (FRLSPI), tunes the weight parameters of the fuzzy-RBF network (a hybrid model constituted by combining Takagi-Sugeno fuzzy rule inference system with RBF network) online and is acquired through combining Least Squares Policy Iteration (LSPI) with fuzzy-RBF network as a function approximator. In FRLSPI, based on the basis functions defined in the fuzzy-RBF network, a solution has been provided for the challenge of determining the state-action basis functions in LSPI. We also provide positive theoretical results concerning an error bound between the optimal and the approximated Action Value Function (AVF) for FRLSPI. Our proposed method has suitable features such as positive mathematical analysis, learning rate independency and, comparatively good convergence properties. Simulation studies regarding the mountain-car control task and acrobat problem demonstrate the applicability and performance of our learning framework. The overall results indicate that the proposed idea can outperform previously known reinforcement learning algorithms.
Decision Making
AMALENDU SI; Sujit Das
Volume 22, Issue 2 , March and April 2025, Pages 81-95
Abstract
The concept of a multi-fuzzy set (MFS) is a hybrid mathematical approachthat aids reasoning and decision-making in situations characterized by imprecise information and multiple occurrences. Researchers have developed robustMFS-based frameworks that facilitate decision-making by identifying optimalalternatives. ...
Read More
The concept of a multi-fuzzy set (MFS) is a hybrid mathematical approachthat aids reasoning and decision-making in situations characterized by imprecise information and multiple occurrences. Researchers have developed robustMFS-based frameworks that facilitate decision-making by identifying optimalalternatives. However, these frameworks often struggle to select the desired alternative in uncommon situations. To address the limitations of existing methods,we incorporated relations and operators with MFS to better measure levels ofuncertainty. To enhance the effectiveness of decision-making, we proposed twoapproaches based on the significance of the criteria within the MFS framework.First, we introduced a relative weight-based approach, where the weight of eachcriterion is estimated dynamically. Second, we developed a normalized-baseddecision-making approach, which generates optimal solutions based on normalized score factors. We demonstrated the effectiveness of our proposed approachesusing semi-realistic cases in health science and healthcare management. Furthermore, we evaluated the performance of the healthcare system across differentsocio-economic regions, which helped to illustrate the relative importance of thecriteria
Fuzzy modelling
Esmaeil Bastami; Hadi Soltanizadeh; Mohammad Rahmanimanesh; Parviz Keshavarzi; Masoud Asadpour
Volume 22, Issue 2 , March and April 2025, Pages 97-127
Abstract
Message-passing graph neural network (MPNN) is a highly e effective learning tool for graph-structured data but is particularly vulnerable to adversarial attacks on information ow, where poisoned data from attacked nodes can propagate to neighboring nodes and be aggregated, amplifying the negative impact ...
Read More
Message-passing graph neural network (MPNN) is a highly e effective learning tool for graph-structured data but is particularly vulnerable to adversarial attacks on information ow, where poisoned data from attacked nodes can propagate to neighboring nodes and be aggregated, amplifying the negative impact across the graph. The malicious perturbations to node features or graph structures can degrade model performance in defensive decision making, which requires a clean ow. We propose a creative defense mechanism to robustly enhance the resilience of MPGNNs against adversarial perturbations by leveraging a creative knowledge-based fuzzy approach for node classification that is implemented by a rule-based fuzzy model. Our study assigns each neighboring node a membership function, which quantifies its relevance to the target node based on knowledge extracted from structural information, node attributes, and community relationships to the target node. By dynamically adjusting the influence of neighboring nodes, the proposed model mitigates the effects of adversarial noise and reduces the impact of irrelevant or poisoned data, ensuring more robust message aggregation. Furthermore, the fuzzy approach helps prevent over-squashing, a common issue when increasing the number of GNN layers, by maintaining important information through controlled message passing and is followed by picking an opinion leader agent from each egonet to recycle the graph. However, our model does not require prior knowledge of the graph or attack types, making it adaptable to various datasets and attack scenarios. Extensive experiments on benchmark datasets, including Cora, PubMed, and Citeseer, demonstrate greater robustness and better accuracy when compared to state-of-the-art defense strategies with reasonable complexity.
Social Sciences
Sajad Bastami; Rojiar Pirmohamadiani; Mohammad bagher Dowlatshahi; Alireza Abdollahpouri
Volume 22, Issue 2 , March and April 2025, Pages 129-146
Abstract
Graph neural networks and fuzzy models offer effective and practical methods for solving various tasks at the large-scale graph level. Large-scale graph embedding based on deep methods and fuzzy models is categorized into fusion and integration. Feature extraction and graph structure at the local and ...
Read More
Graph neural networks and fuzzy models offer effective and practical methods for solving various tasks at the large-scale graph level. Large-scale graph embedding based on deep methods and fuzzy models is categorized into fusion and integration. Feature extraction and graph structure at the local and global levels are based on augmented graph fusion. In fusion-based graph embedding, the fuzzy model is used as an activation function based on an aggregated process. In some cases, the fusion of graph neural network methods with fuzzy systems has been successful. However, no effective methods have been developed for integrating fuzzy models with deep methods. Two main issues are associated with this integration: (1) computational complexity due to the exponential increase in fuzzy rules with the number of features, and (2) the complexity of the solution space due to the combination of fuzzy regression rules between inputs and outputs. Additionally, modeling at the large-scale graph level using linear regression and graph neural networks is not sufficient. Therefore, this paper proposes a feature and structure combination method at the local and global levels using a combination of fuzzy modeling and graph transformers, an integrated deep learning technique called Fuzzy Graph Transformer (FuzzyGT). We conducted experiments on deep learning graph datasets to compare with the proposed model. Our method achieved the best results compared to other advanced models
Engineering
Amir Arsalan Haghrah; Sehraneh Ghaemi; Mohammad Ali Badamchizadeh
Volume 22, Issue 2 , March and April 2025, Pages 147-170
Abstract
Euler-Lagrange systems describe a wide range of mechanical and robotic systems. Uncertainties and dynamic parameterchanges pose significant challenges to control the Euler-Lagrange systems using traditional methods. In suchcases, intelligent methods can cover the limits of the classical techniques. In ...
Read More
Euler-Lagrange systems describe a wide range of mechanical and robotic systems. Uncertainties and dynamic parameterchanges pose significant challenges to control the Euler-Lagrange systems using traditional methods. In suchcases, intelligent methods can cover the limits of the classical techniques. In this research, we intend to present afuzzy logic-based controller trained by the proposed incremental learning algorithm to face the challenges of Euler-Lagrange systems. Incremental learning aims to accumulate experiences over time to train the models. We haveperformed several simulations to test the capabilities of the proposed method and compared the results with wellknownmachine learning-based methods using various criteria. Considering the integral of absolute error, the resultsshow that the proposed method has improved by 40.89%, 38.32%, 34.12%, and 34.79% compared to the best othermethod in nominal system scenario and three other scenarios considering three different levels of uncertainty. Theovershoot of the system response achieved by the proposed control scheme is approximately 44 − 48% less than thebest other method in four scenarios. Also, we have studied the system response to disturbance and noise.
Fuzzy modelling
Johanna Mugisa Atenyi; Chuanjiang Wang
Volume 22, Issue 2 , March and April 2025, Pages 171-186
Abstract
Precise temperature regulation is essential in various industrial applications, particularly in environments requiring high accuracy and stability, such as egg incubation. Conventional control strategies, including Proportional-Integral-Derivative (PID) controllers and Fuzzy Inference Systems (FIS), ...
Read More
Precise temperature regulation is essential in various industrial applications, particularly in environments requiring high accuracy and stability, such as egg incubation. Conventional control strategies, including Proportional-Integral-Derivative (PID) controllers and Fuzzy Inference Systems (FIS), often exhibit limitations in handling nonlinearities, disturbances, and uncertainties. To address these challenges, this research proposes a Fractional Fuzzy Inference System-based PID (FFIS-PID) controller, which enhances the adaptability and robustness of temperature control mechanisms. Unlike traditional fuzzy systems that rely solely on membership degrees, FFIS introduces fractional membership functions and fractional indices, enabling a more flexible and dynamic interpretation of fuzzy rules. The key innovation lies in the fractional compositional rule of inference, which allows the system to intelligently balance the influence of rules by adjusting their impact based on both the truth degree and the information volume. This enhances the adaptability of the control strategy without altering the fundamental rule base structure. The study involves designing fractional membership functions, selecting optimal fractional indices, and evaluating their effects on system behavior. A comparative analysis between FIS-PID and FFIS-PID controllers is conducted through simulations and experimental validation on an incubator system. The results confirm that the FFIS-PID controller provides superior temperature regulation by enabling real-time adaptability to changing conditions. This work contributes to the field of intelligent control by providing a novel approach to fuzzy inference enhancement through fractional compositional rule of inference mechanism. Future research could extend this methodology to other nonlinear control applications, further leveraging fractional indices for improved decision-making and stability.
Fuzzy approximation
Mohadese Rezaei; Hossein NezamAbAdipour
Volume 22, Issue 2 , March and April 2025, Pages 187-204
Abstract
Portfolio management is a challenging task due to the uncertainty and volatility in financial markets, making precise asset allocation and return maximization difficult. This paper presents a novel deep reinforcement learning (DRL) approach enhanced with fuzzy trend indicators to improve portfolio decision-making. ...
Read More
Portfolio management is a challenging task due to the uncertainty and volatility in financial markets, making precise asset allocation and return maximization difficult. This paper presents a novel deep reinforcement learning (DRL) approach enhanced with fuzzy trend indicators to improve portfolio decision-making. The model was developed using a DRL framework, where a convolutional neural network (CNN)-based policy network learns to optimize asset allocations through interactions with the market. Fuzzy trend indicators are incorporated as additional input features, enabling the model to better capture market uncertainties and ambiguous trends. By providing a more flexible representation of market conditions, fuzzy trend indicators allow the model to dynamically adjust portfolio allocations in response to changing trends, leading to more precise asset allocation decisions and enhanced portfolio performance. The proposed model was trained and evaluated on historical stock data from the Brazilian stock market, covering the period from 2011 to 2020. The dataset includes daily high, low, and closing prices, ensuring a strong foundation for model training and validation. Experimental results show that the fuzzy-enhanced model outperforms some state-of-the-art strategies in terms of both returns and adaptability to volatile market conditions.
