Fuzzy Neural Network Model Research Articles

An Evolving Fuzzy Neural Network Based on Or-Type Logic Neurons for Identifying and Extracting Knowledge in Auction Fraud

The rise in online transactions for purchasing goods and services can benefit the parties involved. However, it also creates uncertainty and the possibility of fraud-related threats. This work aims to explore and extract knowledge of auction fraud by using an innovative evolving fuzzy neural network model based on logic neurons. This model uses a fuzzification technique based on empirical data analysis operators in an evolving way for stream samples. In order to evaluate the applied model, state-of-the-art neuro-fuzzy models were used to compare a public dataset on the topic and, simultaneously, validate the interpretability results based on a common criterion to identify the correct patterns present in the dataset. The fuzzy rules and the interpretability criteria demonstrate the model’s ability to extract knowledge. The results of the model proposed in this paper are superior to the other models evaluated (close to 98.50% accuracy) in the test.

A self-organizing fuzzy neural network modeling approach using an adaptive quantum particle swarm optimization

A self-organizing fuzzy neural network modeling approach using an adaptive quantum particle swarm optimization

Real-Time Regulation of Physical Training Intensity Based on Fuzzy Neural Network

In this paper, the fuzzy neural network model is studied, the real-time regulation model of physical training intensity is analyzed and a real-time regulation system based on a fuzzy neural network is designed. The real-time, accurate and effective regulation of the physiological load intensity in the body of the exerciser is consistent with the predetermined goals of the training program. In this paper, we propose an RBF neural network, combined with the plan and demand of physical training operation situation sensing, and considering that most of the biological training operation data is fuzzy, this paper connects a fuzzy logic inference system and a neural network and proposes a network operation situation sensing model based on an RBF neural network structure. The RBF neural network and the traditional fuzzy neural network are compared. The experiments prove that this paper’s fuzzy neural network model has a faster training speed. In this paper, we use time-realistic control equipment to monitor the physical training process of athletes so that we can grasp the training situation of athletes in real-time and ensure that athletes can achieve better training results by changing training methods and changing training loads in time for those athletes who cannot reach their sports goals. In the process of physical fitness training monitoring, an effective monitoring of training, time-accurate regulation monitoring has the advantage of timely feedback on the training situation. This model has a better convergence effect during exercise and a higher accuracy of posture prediction during testing.

Fixed/Preassigned-time synchronization of high-dimension-valued fuzzy neural networks with time-varying delays via nonseparation approach

For the high-dimension-valued fuzzy neural networks (HDVFNNs) models with time-varying delays, the fixed/preassigned-time synchronization is explored deeply in this paper. Instead of the separation approach, a direct method in complex field or quaternion skew field is fully utilized, which is suitable for a wider range. Furthermore, two fuzzy controllers without sign function based on the direct method are designed, which can effectively avoid the frequent switching and chattering phenomena involved by sign functions. Then, we derive two sufficient conditions to realize the desired the fixed/preassigned-time synchronization of the addressed HDVFNNs. Finally, the availability of notional researches has been testified by two numerical simulations.

HFGNN-Proto: Hesitant Fuzzy Graph Neural Network-Based Prototypical Network for Few-Shot Text Classification

Few-shot text classification aims to recognize new classes with only a few labeled text instances. Previous studies mainly utilized text semantic features to model the instance-level relation among partial samples. However, the single relation information makes it difficult for many models to address complicated natural language tasks. In this paper, we propose a novel hesitant fuzzy graph neural network (HFGNN) model that explores the multi-attribute relations between samples. We combine HFGNN with the Prototypical Network to achieve few-shot text classification. In HFGNN, multiple relations between texts, including instance-level and distribution-level relations, are discovered through dual graph neural networks and fused by hesitant fuzzy set (HFS) theory. In addition, we design a linear function that maps the fused relations to a more reasonable range in HFGNN. The final relations are used to aggregate the information of neighbor instance nodes in the graph to construct more discriminative instance features. Experimental results demonstrate that the classification accuracy of the HFGNN-based Prototypical Network (HFGNN-Proto) on the ARSC, FewRel 5-way 5-shot, and FewRel 10-way 5-shot datasets reaches 88.36%, 94.45%, and 89.40%, respectively, exceeding existing state-of-the-art few-shot learning methods.

Meticulous Land Cover Classification of High-Resolution Images Based on Interval Type-2 Fuzzy Neural Network with Gaussian Regression Model

This paper proposes a land cover classification method that combines a Gaussian regression model (GRM) with an interval type-2 fuzzy neural network (IT2FNN) model as a classification decision model. Problems such as the increase in the complexity of ground cover, the increase in the heterogeneity of homogeneous regions, and the increase in the difficulty of classification due to the increase in similarity in different regions are overcome. Firstly, the local spatial information between adjacent pixels was introduced into the Gaussian model in image gray space to construct the GRM. Then, the GRM was used as the base model to construct the interval binary fuzzy membership function model and characterize the uncertainty of the classification caused by meticulous land cover data. Thirdly, the upper and lower boundaries of the membership degree of the training samples in all categories and the principle membership degree as input were used to build the IT2FNN model. Finally, in the membership space, the neighborhood relationship was processed again to further overcome the classification difficulties caused by the increased complexity of spatial information to achieve a classification decision. The classical method and proposed method were used to conduct qualitative and quantitative experiments on synthetic and real images of coastal areas, suburban areas, urban areas, and agricultural areas. Compared with the method considering only one spatial neighborhood relationship and the classical classification method without a classification decision model, for images with relatively simple spatial information, the accuracy of the interval type-2 fuzzy neural network Gaussian regression model (IT2FNN_GRM) was improved by 1.3% and 8%, respectively. For images with complex spatial information, the accuracy of the proposed method increased by 5.0% and 16%, respectively. The experimental results prove that the IT2FNN_GRM method effectively suppressed the influence of regional noise in land cover classification, with a fast running speed, high generalization ability, and high classification accuracy.

Features of Methods for Assessing the Long-Term Traffic Intensity in the Design of Toll Roads

Long-term traffic intensity is among core characteristics that determine main parameters while developing projects for construction of new highways. The long-term traffic intensity influences estimated speed, pavement design, total number of traffic lanes, width of traffic lanes and roadsides, longitudinal slope, radii of horizontal curve, transverse slope, radii of convex and concave curves regarding the longitudinal profile, width of median strip, layout of intersection or junction with other roads.Existing methods for predicting traffic intensity for toll roads are also deterministic and cannot estimate the range of values for the listed indicators. In this regard, the objective of the study is to identify the features, advantages, and disadvantages of existing methods for assessing the long-term traffic intensity for toll roads.The study considered both traditional, classical methods (extrapolation, historical analysis, approximation) and promising innovative approaches based on the theory of fuzzy logic and neural network modelling.

Rolling force prediction in cold rolling process based on combined method of T-S fuzzy neural network and analytical model

Rolling force prediction in cold rolling process based on combined method of T-S fuzzy neural network and analytical model

Learning Fuzzy Neural Networks by Using Improved Conjugate Gradient Techniques

One of the optimal approaches for learning a Takagi Sugeno-based fuzzy neural network model is the conjugate gradient method proposed in this research. For the PRP and the LS approaches, a novel algorithm based on the Liu-Storey (LS) approach is created to overcome the slow convergence. The developed method becomes descent and convergence by assuming some hypothesis. The numerical results show that the developed method for classifying data is more efficient than the other methods, as shown in Table (2), where the new method outperforms the others in terms of average training time, average training accuracy, average test accuracy, average training MSE, and average test MSE.

Research on Image Texture Feature Extraction Based on Digital Twin

The purpose of image smoothing is to improve the visual effect of the image and improve the clarity of the image, so as to make the image more conducive to computer processing and various feature analysis. Because the current technology fails to smooth the preprocessed image, it leads to the extraction of image texture features. The anti-interference performance is weak. For this reason, an image texture feature extraction technology based on the digital twin is proposed. Similarity analysis is carried out through the internal structure of the image, and the image is smoothed by the semisupervised learning method. On the basis of optimizing the denoised image through digital twinning, detect target feature points in the original image, then remove the abnormal and split feature points, assign the direction of image texture feature points, and build a fuzzy back propagation neural network model. Image texture feature extraction technology is implemented. The experimental results show that, compared with the traditional method, the proposed technique has a strong identification of original image features, and has a strong consistency with original data, and has a strong ability to resist the influence of abnormal data, noise, or redundant feature points.

Nonstationary fuzzy neural network based on FCMnet clustering and a modified CG method with Armijo-type rule

Nonstationary fuzzy inference systems (NFISs) model the variation in opinions of individual experts and expert groups. They have the capability similar to type-2 fuzzy systems in some regards with less computational costs. This paper presents a nonstationary fuzzy neural network (NFNN) model by combining NFISs and neural networks (NNs). The proposed model synthesizes the logical inference and language expression abilities of NFIS with the learning mechanism of NNs. Hence, it makes NNs more translatable and also achieves self-learning of fuzzy rules. Besides, a fuzzy c-means network (FCMnet) clustering and a modified conjugate gradient method with constrained Armijo-type rule (MCGA) are proposed to initialize and train NFNN, respectively. The proposed FCMnet first combines the fuzzy c-means (FCM) algorithm with NNs. Thus, it not only improves the performance of FCM, but also makes NNs more interpretable. For MCGA, we construct a specific conjugate coefficient to ensure the sufficient descent property and propose a constrained Armijo-type rule to search a suitable learning rate in each iteration. By adopting these two techniques, MCGA achieves a fast convergence speed. In addition, both weak and strong convergence results are proven rigorously. Experiments on 14 datasets are carried out to illustrate the competitive performance of our models.

Adaptive decision model of product team organization pattern for extracting new energy from agricultural waste

The objective of this work is to solve the low efficiency in the extraction of new energy from Agricultural Waste (AW). In this work, the development and utilization of AW are discussed in depth by combining with the current technological progress. To effectively implement new energy mining projects, the adaptive decision model of product team organization pattern is studied deeply, and the Fuzzy Neural Network (FNN) model is mainly used as the decision method to select product team organization pattern. The biomass waste is solidified and formed as fuel. The principle is to crush biomass waste to a certain extent and compress it into rod or block by physical extrusion. During training of the FNN, each parameter needs to be constantly adjusted to make the new data reach the target value fitting output value as far as possible after passing the FNN. The test results show that the FNN has achieved satisfactory prediction results in 18 groups of test data. The point structure entropy and edge structure entropy of product innovation service team are −1.931 and 0.09 respectively. To sum up, extracting new energy from AW can realize sustainable development of society. The adaptive decision model based on product team organization pattern can complete the effective connection of energy product constituent entities, provide technical support for the maximum utilization of social service resources, and promote the extraction and utilization of new energy more efficiently.

An expert system for insect pest population dynamics prediction

Avocado (Persea americana) production is increasing in Kenya, with both small and largeholder farming for domestic and export markets. However, one of main challenges that limit production is infestation by insect pests, notably the oriential fruit fly Bactocera dorsalis and Ceratitis spp. fruit flies, which cause direct crop losses and are indirectly responsible for non-tariff trade barriers due to stringent export requirements. Data on weekly pest trap counts were collected between September 2017 and December 2020 within orchards in avocado plantations. Fuzzy neural network (FNN) were used to model the population dynamics of B. dorsalis and Ceratitis spp. Weekly pest counts, rainfall, average temperature, relative humidity and avocado plant physiological stages were used for predictive modeling in different orchards. The performance of the resulting models was evaluated using coefficient of determination (R2), mean absolute error (MAE), mean relative approximation error (MRAE) and root mean squared error (RMSE). FNN models achieved satisfactory results in predicting the dynamics of the pests in the orchards, with most of the models obtaining R2 > 0.85. We demonstrated how FNN models can be used as predictive tools for managing and controlling fruit fly pest populations in these plantations, and how they may be suitable to predict fruit fly or other pests in similar cropping systems. Once the input variables are known, they can be loaded into the FNN models to predict field pest populations, and based on threshold values, allow for implementation of timely and adequate control measures such as the use of biopesticides.

Research on the Multimodal Digital Teaching Quality Data Evaluation Model Based on Fuzzy BP Neural Network.

We propose in this paper a fuzzy BP neural network model and DDAE-SVR deep neural network model to analyze multimodal digital teaching, establish a multimodal digital teaching quality data evaluation model based on a fuzzy BP neural network, and optimize the initial weights and thresholds of BP neural network by using adaptive variation genetic algorithm. Since the BP neural network is highly dependent on the initial weights and points, the improved genetic algorithm is used to optimize the initial weights and thresholds of the BP neural network, reduce the time for the BP neural network to find the importance and points that satisfy the training termination conditions, and improve the prediction accuracy and convergence speed of the neural network on the teaching quality evaluation results. The entropy value method, a data-based objectivity evaluation method, is introduced as the guidance mechanism of the BP neural network. The a priori guidance sample is obtained by the entropy method. Then, the adaptive variational genetic algorithm is used to optimize the BP neural network model to learn the a priori sample knowledge and establish the evaluation model, which reduces the subjectivity of the BP neural network learning sample. To better reflect and compare the effects of the two neural network evaluation models, BP and GA-BP, the sample data were continued to be input into the original GA and BSA to obtain the evaluation results and errors; then, the evaluation results of the two evaluation models, BP and GA-BP, were compared with the evaluation results of the two algorithms, GA and BSA. It was found that the GA-BP neural network evaluation model has higher accuracy and can be used for multimodal digital teaching quality evaluation, providing a more feasible solution.

The Optimization of Hospital Financial Management Based on Cloud Technology and Wireless Network Technology in the Context of Artificial Intelligence

In the reign of Internet technology, cloud technology has emerged as one of the powerful tools in computing with many advantages in the bag. Cloud technology is defined as online access to all types of computing applications. The applications range from access to data, servers, software, databases to storage and retrieval of data. The data can be accessed from anywhere, anytime. The user does not need to own the hardware in this case. Clouds have multiple data centers located at various places. On-demand data service is the best part of cloud technology. This technology is made use in many industries. In this research, we will study how cloud technology and wireless networking technology are applied to optimize hospital financial management. The healthcare industry faces various challenges in financial management such as budgeting, growth planning, and cost-effectiveness. Considering this scenario, it is evident that there is a solid need to optimize its financial management. The cloud computing technology driven by artificial intelligence (AI) is deployed in carrying out repetitive tasks and automation, resulting in increased productivity. In this research, the proposed system uses wireless networking systems and cloud technology to optimize hospital financial management. The model works with the implementation of a hybrid of fuzzy neural networks. The proposed model is analyzed with the existing fuzzy logic and neural network models, and it is found that the proposed model has obtained an accuracy of 97.56% with minimum resources for successful financial management at hospitals.

The Construction of Virtual Experiment Platform Based on the Fuzzy System

Virtual experiment platform is an open network virtual experiment system, and it is constructed by virtual reality technology. It virtualizes and digitizes experimental resources in the current virtual experiment platform. Virtual experimental platform is similar to real experimental platform. Users can adjust the virtual instrument to complete the experiment process. In this paper, a virtual experiment platform model of the fuzzy system is built. Based on the acquired correlative variables of the fuzzy system, a virtual experiment platform model of the fuzzy system based on the recursive fuzzy network is established. And the model is trained and modified by using the running data of the actual fuzzy algorithm, and the prediction of the fuzzy system in the process of virtual experiment platform is realized. The experimental results show that the virtual experiment platform model of the fuzzy system based on the recursive fuzzy neural network can obtain higher prediction accuracy. Compared with standard fuzzy neural network and other modeling methods, the model of virtual experiment platform of the fuzzy system proposed in this paper can obtain better prediction effect. So, the virtual experiment platform based on the fuzzy algorithm is feasible. The virtual experiment simulation platform of the fuzzy system is built, and the experiment design and research are carried out through the virtual experiment simulation platform.

T036 Classification of breast cancer stadium using fuzzy neural network (FNN) model algorithm

T036 Classification of breast cancer stadium using fuzzy neural network (FNN) model algorithm

Study on Dynamic Stability Prediction Model of Slope in Eastern Tibet Section of Sichuan-Tibet Highway

The stability of the slope along the middle section of Tibet controls the safety and smoothness of the Sichuan-Tibet highway, which is affected by multiple and uncertain factors such as rainfall. The slope dynamic stability is evaluated to the benefit of that salvager can prepare in advance and preserve timely and accurately. Therefore, engineering treatment scheme in different batches, stages, and grades can be proposed prospectively. Random Forest algorithm was used to rank 10 primary factors: precipitation, earthquake, human factors, groundwater, slope height, slope gradient, dense degree, weathering depth, vegetation, and slope shape. Considering precipitation and earthquake as dynamic factors, a wavelet and NARX dynamic neural network were used to predict the trend and quantity of precipitation and earthquake, followed by developing a dynamic stability evaluation model by combining a fuzzy neural network model with other indexes. Results show that (1) the superposition error in rainfall and earthquake prediction is 0.21%, proving that the ranking of influencing factors is reasonable, and (2) the back-judgment and test accuracy of the dynamic evaluation model are 93.98% and 91.67%, respectively, indicating that the model is accurate and applicable. The model can evaluate the dynamic stability of slopes and provide more reasonable engineering protection countermeasures so that Highway Public Works Department can deal with emergencies and disasters timely and precisely.

Incorporating Physical Models for Dynamic Stall Prediction Based on Machine Learning

Unsteady aerodynamic prediction is an important part of modern aircraft safety and control law design. To improve the accuracy and efficiency for unsteady aerodynamic prediction of aircraft at high angles of attack, this paper proposed a machine learning framework based on multifidelity methods. The framework combines the linear dynamic derivative model and the fuzzy neural network model, which can achieve higher prediction accuracy under sparse experimental states. A series of wind-tunnel tests was carried out for the pitching motions of NASA Common Research Model at high angles of attack, to obtain steady and unsteady aerodynamic loads. These experimental data are used to verify the prediction accuracy of the unsteady model in a wide range of oscillation amplitude, frequencies, and mean angles of attack. The results show that the method has good generalization capability for the parameters of interest. At the same time, the comparison with the prediction results only from high-fidelity data shows that the proposed method can effectively reduce the amount of data required for the model of training and improve the modeling robustness to different types of motions.

Modeling and simulation analysis of electric forklift energy prediction management

In recent years, the effective utilization of energy for electric forklift has been the focus of research. However, the load of forklift change markedly and frequently, leading to a great challenge in the performance of the composite energy system. To solve this problem, this paper presents the power system structure of electric forklift and the battery–supercapacitor hybrid energy management method of electric forklift truck. The working conditions and energy flow of electric forklift are analyzed and the power demand forecasting model of electric forklift power system is indispensable. Two predictive control based on Markov chain and neural network are developed, considering the performance of predictive control. The predictive model is applied to the stochastic model predictive (SMPC) energy management strategy. Simulations are carried out on the representative driving cycles of an electric forklift. As an auxiliary power supply on the energy management of composite power, the influence of the development depth of supercapacitor is discussed. Simulation results show that short-term Markov chain model is closer to the actual working condition. Fuzzy neural network model can give full play to the charging and discharging characteristics of supercapacitors. The state of charge of supercapacitor is stable between 0.5 and 1, which effectively stabilizes the rate of the charge and discharge of the battery.