Fuzzy Selector Research Articles

UNFIS: A Novel Neuro-Fuzzy Inference System with Unstructured fuzzy rules

An important constraint of Fuzzy Inference Systems (FIS) is their structured rules. Indeed, each rule should evaluate all input variables in its antecedent part. Thus, the length of all fuzzy rules and the number of input variables are equal. However, in many decision-making problems evaluating some conditions on a limited set of input variables is sufficient to decide properly (unstructured rules). Consequently, the FIS’s performance, generalizability, and interpretability are restricted by the aforementioned limitation. This study provides a neuro-fuzzy inference system that can create each fuzzy rule taking into account various sets of input variables in order to handle this problem. To realize this capability, a new fuzzy selector neuron with an adaptive parameter is proposed that can select input variables in the antecedent part of each fuzzy rule. Moreover, the consequent part of the Takagi–Sugeno–Kang FIS is properly changed to consider only the selected set of input variables. To learn the parameters of the proposed architecture, the Levenberg–Marquardt (LM) method is used for regression problems. For classification applications, a trust-region-based learning method (General quasi-Levenberg–Marquardt (GqLM)) is proposed to minimize cross-entropy in multiclass problems. The performance of the proposed method is compared with some related previous approaches in some real-world classification and regression problems. Based on these comparisons, the proposed method has better or very close performance with a parsimonious structure consisting of unstructured fuzzy rules.

A LLM-Based Hybrid-Transformer Diagnosis System in Healthcare.

The application of computer vision-powered large language models (LLMs) for medical image diagnosis has significantly advanced healthcare systems. Recent progress in developing symmetrical architectures has greatly impacted various medical imaging tasks. While CNNs or RNNs have demonstrated excellent performance, these architectures often face notable limitations of substantial losses in detailed information, such as requiring to capture global semantic information effectively and relying heavily on deep encoders and aggressive downsampling. This paper introduces a novel LLM-based Hybrid-Transformer Network (HybridTransNet) designed to encode tokenized Big Data patches with the transformer mechanism, which elegantly embeds multimodal data of varying sizes as token sequence inputs of LLMS. Subsequently, the network performs both inter-scale and intra-scale self-attention, processing data features through a transformer-based symmetric architecture with a refining module, which facilitates accurately recovering both local and global context information. Additionally, the output is refined using a novel fuzzy selector. Compared to other existing methods on two distinct datasets, the experimental findings and formal assessment demonstrate that our LLM-based HybridTransNet provides superior performance for brain tumor diagnosis in healthcare informatics.

Classification of skin cancer stages using a AHP fuzzy technique within the context of big data healthcare.

Skin conditions in humans can be challenging to diagnose. Skin cancer manifests itself without warning. In the future, these illnesses, which have been an issue for many, will be identified and treated. With the rapid expansion of big data healthcare framework summarization and precise prediction in early stage skin cancer diagnosis, the fuzzy AHP technique produces the best results in both of these fields. Big data is a potent technology that enhances the standard of research and generates better results more rapidly. This essay gives a way to group the stages of skin cancer treatment based on this information. The combination of support vector machine multi-class classification and fuzzy selector with radial basis function-based binary migration classification of virtual machines is put through a number of experiments. The connections have been categorized. These examinations have determined whether the tumors are malignant or benign and how malignant they are. The images of spots on the skin acquired from laboratory images make up the data set used for processing. We have talked about how to handle and process large datasets in the area of classification using MATLAB, like skin spot images. Our technique outperforms competing approaches by maintaining stability even as the size of the data set grows rapidly and with little error. In comparison to other methods, the suggested approach meets the accuracy criterion for correct classifications with a score of 90.86%. As a result, the proposed solution is viewed as a potentially useful tool for identifying mass stages and categorizing skin cancer severity.

Trajectory tracking of redundantly actuated mobile robot by MPC velocity control under steering strategy constraint

Four-wheel independent-steering redundantly-actuated omnidirectional mobile robot (FIR-OMR) has attracted wide attention for its excellent motion performance such as multi-mode steering and flexible attitude adjustment. However, for the 8-drive redundant and strong nonlinear system of FIR-OMR, the complexity and time-consuming of model solutions make it hard for the embedded system to achieve fast and efficient trajectory tracking. To copy with that, an optimal velocity model predictive control under steering strategy constraint (USSC-MPC) is presented, which is combined with pose correction by local micro error (LME-PC) for independent wheel angle feedback control. Specifically, the steering modes of FIR-OMR are divided into fixed independent wheel steering (IWS), zero-angle pure differential steering (DS), and Ackerman steering (AS), which is ultimately selected by steering fuzzy selector (SFS). Then a steer-based model prediction controller is designed to optimize the wheel velocity and the micro-error feedback wheel angle of pose correction pointing to the desired point is calculated by FIR-OMR in real-time. As a result, by controlling the wheel velocity and wheel angle, a satisfactory path tracking control effect can be obtained. In our proposed method, the USSC-MPC ensures the optimality of wheel velocity, and the LME-PC can achieve efficient trajectory tracking with a faster error convergence rate. Through simulations and experiments, complex motions such as multi-section paths and side parking can be realized, which verifies the feasibility and superiority of our method.

Novel hybrid fuzzy-PID control scheme for air supply in PEM fuel-cell-based systems

This paper proposes a novel hybrid fuzzy-PID controller for air supply on Proton Exchange Membrane fuel cell (PEMFC) systems. The control objective is to adjust the oxygen excess ratio at a given setpoint in order to prevent oxygen starvation and damage of the fuel-cell stack. The proposed control scheme consists of three parts: a fuzzy logic controller (FLC), a fuzzy-based self-tuned PID (FSTPID) controller and a fuzzy selector. Depending on the value of the error between the current value of oxygen excess ratio and its setpoint value, the fuzzy selector decides which controller should play the greatest effect on the control system. The performance of the proposed control strategy is analysed through simulations for different load variations and for parameter uncertainties. The results show that the novel hybrid fuzzy-PID controller performs significantly better than the classical PID controller and the FLC in terms of several key performance indices such as the Integral Squared Error (ISE), the Integral Absolute Error (IAE) and the Integral Time-weighted Absolute Error (ITAE), as well as the overshoot, settling and rise time for the closed-loop control system.

Reducing the size of traveling salesman problems using vaccination by fuzzy selector

At present, no algorithm can solve Combinatorial Optimization Problems (COPs) in polynomial time. The Traveling Salesman Problem (TSP) is an NP-hard COP example widely used as a test problem since it is mathematically equivalent to others in diverse fields. State of the art algorithms provide low quality solutions for large instances at high computational cost. Reducing the TSP size is a heuristic that works well finding solutions at a reduced cost. The Reduce-Optimize-Expand (ROE) methodology allows to solve COPs faster by improving the quality of solutions, however, obtaining optimal reductions is still an open problem since there is no way to determine if the removed nodes are part of the optimal solution without knowing it beforehand. This paper presents an intelligent strategy with a fuzzy logic classifier, based on the ROE, to obtain systematic reductions of TSP instances. The method was tested using TSP instances from the TSPLIB, ranging from 131 to 2924 cities. Comparative experimental results obtained higher-quality reductions, thus improving the overall performance of ROE when using the intelligent selection strategies. Intelligent reductions contribute to improve the performance of existing TSP algorithms, even given solution to many unsolved problems with huge amount of nodes by reducing them to a treatable size.

Reducing the Size of Combinatorial Optimization Problems Using the Operator Vaccine by Fuzzy Selector with Adaptive Heuristics

Nowadays, solving optimally combinatorial problems is an open problem. Determining the best arrangement of elements proves being a very complex task that becomes critical when the problem size increases. Researchers have proposed various algorithms for solving Combinatorial Optimization Problems (COPs) that take into account the scalability; however, issues are still presented with larger COPs concerning hardware limitations such as memory and CPU speed. It has been shown that the Reduce-Optimize-Expand (ROE) method can solve COPs faster with the same resources; in this methodology, the reduction step is the most important procedure since inappropriate reductions, applied to the problem, will produce suboptimal results on the subsequent stages. In this work, an algorithm to improve the reduction step is proposed. It is based on a fuzzy inference system to classify portions of the problem and remove them, allowing COPs solving algorithms to utilize better the hardware resources by dealing with smaller problem sizes, and the use of metadata and adaptive heuristics. The Travelling Salesman Problem has been used as a case of study; instances that range from 343 to 3056 cities were used to prove that the fuzzy logic approach produces a higher percentage of successful reductions.

A Neurofuzzy Structure Modelling Evapotranspiration in a Greenhouse Crop

This paper shows the applicability of a neurofuzzy method applied to a tomato plant (Lycopersicon esculentum, Mill), in order to obtain a model of the instantaneous evapotranspiration. Two operational dynamics (diurnal and nocturnal) are defined in a hierarchical fuzzy model by a Takagi-Sugeno (T-S) type with linear consequents. The fuzzy selector of the two dynamics is the solar radiation measure. The fuzzy C-means algorithm is used to identify the fuzzy rules premises. The hierarchical fuzzy modelling is used to obtain a multi-model of the evapotranspiration cycles. In order to simplify the model structure, the variable of vapour pressure deficit (VPD) is introduced, and thus, a physical interpretation of the interior climate is obtained. VPD helps to preserve the quality and production level in the greenhouse crop.