Use Of Fuzzy Neural Networks Research Articles

Нейромережеві прогностичні технології класифікації криз та оцінки банкрутства суб’єктів торговельної галузі

The ongoing war in Ukraine has led to a significant increase in the number of business bankruptcies. Timely assessment of bankruptcy risk is critical for making effective management decisions. The aim of the study is to develop an adaptive model for assessing the propensity of an enterprise to bankruptcy based on fuzzy neural networks, taking into account the specifics of the Ukrainian market in wartime. To achieve the aim, several stages were carried out: formation of a sample of enterprises in the trade industry; determination of crisis classes using cluster analysis; building a fuzzy neural network using the Fuzzy Logic Toolbox package in MATLAB; training the network on historical data and testing on new data. At the first stage, a sample of enterprises is formed, with the help of which the model will be trained. At the second stage, classes of crisis (threat of bankruptcy) are formed. At the third stage, with the help of the apparatus of fuzzy neural networks, a membership function with a certain set of parameters is formed. With the help of this function, it is possible to attribute this or that enterprise with a certain degree of belonging to one of the classes of crisis formed at the previous stage. The fourth stage is to check the adequacy of the obtained model and the possibility of its practical use. At the fifth stage, the practical implementation of the model is carried out: with the help of the membership function, the propensity of the researched enterprise to bankruptcy is determined, that is, to what extent the studied enterprise belongs to one or another class of crisis. As a result of the application of the proposed algorithm, a fuzzy neural network was built, which used the financial indicators of Ukrainian enterprises. To train the model, a dataset was used that included 12 enterprises in the trade industry. The developed model allows to classify enterprises by the level of financial sustainability with high accuracy. Fuzzy neural networks have been shown to be an effective tool for predicting bankruptcy under uncertainty. The use of fuzzy neural networks made it possible to take into account uncertainty and adapt to changes in the external environment. The obtained results indicate the perspective of using fuzzy neural networks to assess the financial sustainability of enterprises. The developed model can be used as a tool for early detection of signs of crisis and making timely managerial decisions.

On the use of fuzzy neural networks in the framework of a risk-based approach in control and supervisory activities in civil aviation

A risk-oriented approach implemented in conducting control and supervisory activities in Civil Aviation organizations makes it possible to increase the effectiveness of such activities, the objectivity of assessments, to reduce costs and the additionalburden on business. The main provisions, regulating the activities of control and supervision bodies, including the issues of risk assessment, are generally specified in regulatory documents. However, uncertainty remains regarding the use of so-called risk indicators, which are designed to forecast risks for flight safety. Currently, there are no guidelines on the number and composition of such indicators, there are no methods to use them for the intended purpose. The article proposes a solution to this problem using elements of artificial intelligence. Based on the example of risk indicators distinctive for air traffic service organizations, the feasibility of forecasting the level of risk through a fuzzy (hybrid) neural network is shown. As is well known, such hybrid structures, combining neural networks and fuzzy logic, collect the best properties of both methods. The formation of a set of risk ndicators and initial data for network training is carried out with the involvement of qualified experts with extensive experience in flight safety management and control and supervisory activities. The trained network allows us to quantify a forecasted level of risk in an airline based on the identified risk indicators considering the degree of their manifestation. All the stages of building and using the network in the ANFIS editor of the MATLAB software package are shown. The proposed method can also be used in the flight safety management systems for various providers of aviation services.

Intellectual Scenario-synergetic Control of the Humidity and Temperature Regime of the Greenhouse Facilities

The article substantiates the management of the humidity and temperature regime of greenhouse complexes on the basis of a scenario-synergetic approach. The scenarios for controlling the temperature and humidity conditions in the greenhouse using the approach of fuzzy neural networks are formed. The structure of an automated control system for technological processes is developed, which provides automated collection and processing of information for the implementation of control actions in order to improve the efficiency of the greenhouse complex on the basis of a scenario-synergetic approach. The corresponding fuzzy neural networks are synthesized for a synergistic assessment of the interaction of technological parameters. Estimation of the root-mean-square error in the synthesis of fuzzy neural networks confirms the possibility of their use for the synergistic formation of scenarios for controlling the temperature and humidity regime in greenhouses to reveal the presence of a synergistic effect. Production rules for scenario management of temperature and humidity conditions are formed. It is shown that the use of fuzzy neural networks for the formation of scenarios for controlling the humidity and temperature regime provides the possibility of obtaining the appropriate scenarios for making managerial decisions and their prompt correction.

Neuro-fuzzy system for solving fuzzy singular perturbation problems

In this paper we have introduced a new method, by using fully fuzzy neural network (FFNN) to find the numerical solution for solving fuzzy singular perturbation problems (FSPPS) for ordinary differential equations. It uses the Levenberg-Marquardt training machine algorithm (TrianLM) for calculating the Fuzzy neural network parameters and the sigmoid function of the hidden units is hyperbolic tangent activation function, that is: . The fuzzy trial solution of FSPPs is written as a sum of two parts. The first part satisfies the fuzzy conditions, it contains no fuzzy adjustable parameters. The second part involves a feed-forward fuzzy neural network containing fuzzy adjustable parameters. In comparison with existing numerical methods, shows upp that the use of fuzzy neural networks provides solutions with good generalization and high accuracy. The proposed method is illustrated by several examples.

Technical and economic optimization of fuel facilities – energy complex using elements of fuzzy logic and artificial neural networks

This article discusses the possibilities of optimizing the fuel and energy complex by improving the accuracy of the forecast of the state in both the operational and medium-term periods: from an hour to several years. The model used to implement the mentioned energy efficiency improvement is based on the use of fuzzy neural networks together with the function of minimizing power losses in the elements of electric power systems. The Mamdani modification is used, which takes into account the database of previous periods of the system state, represented as static load characteristics. The organization of this approach allows you to take into account weakly formalized factors, thereby improving the quality of forecasting and dispatching management. At the same time, it is possible to use models for a wide class of objects, which is achieved due to its scalability. The estimation of forecasting accuracy of the proposed approach exceeds similar indicators of currently used statistical methods and regression implementations. These factors are direct economic drivers of reducing production costs.

Analysis of fuzzy logic methods for forecasting customer churn

The object of research is the process of predicting the churn of customers of telecommunications companies based on fuzzy logic and neural networks. The research carried out is based on the application of an approach that is implemented through the combined use of fuzzy logic and neural networks. The main assumption of the study is the hypothesis that the use of a fuzzy neural network formed on the basis of fuzzy logic algorithms can improve the accuracy of predicting customer churn relative to available solutions. This result can’t be achieved neglecting the existing resource constraints and requirements, which must be determined separately for each case of research. The relevance of the problem of forecasting customer churn for companies with a large number of users is considered. A model for predicting customer churn is proposed based on the combined use of fuzzy logic and neural networks. The main feature of this approach is that a test sample of normalized data is used at the basis of fuzzy neural networks, which are processed to form the parameters of membership functions that correspond to the inference system, that is, conclusions are made on the basis of a fuzzy logic apparatus. Also, to find the parameters of the membership function, neural network algorithms are used. Such systems can use previously known information, learn, gain new knowledge, predict time series, perform image classification, and besides, they are quite visual to the user. The application of methods of fuzzy logic is considered, they make it possible to obtain a result in the form of a fuzzy inference. The expediency of choosing these methods is explained by the fact that they were previously used in fuzzy automatic control systems and showed sufficiently high quality results. The expediency and prospects of using the proposed approach in the problem of predicting the outflow of customers of telecommunications companies are shown, and the results of software implementation are presented.

Synthesis of an Expert System for Assessing the Security of Computer Networks Based on a Fuzzy Neural Network

The aim of the article is to substantiate the principles of synthesis of an expert system for assessing the security of computer networks based on a fuzzy neural network, and this is an urgent scientific and technical task. Requirements for the operative security assessment of computer networks for data protection are analyzed. It was shown that data security should be provided by the network administrator or persons who need to use special decision support systems in assessing the security of computer networks. To solve this problem, factors that characterize the security of electronic systems, including computer systems, have been identified; the use of fuzzy neural networks is proposed as a mathematical apparatus for constructing an expert system; a technique for the synthesis of a fuzzy neural network for assessing the security of computer networks has been developed; an appropriate fuzzy neural network has been created and tested for adequacy; the prospects of the proposed methodology for creating an expert system for assessing the security of computer systems have been established. The scientific and practical significance of developing such a system lies in the fact that a fuzzy neural network is configured on a specific object in order to quickly determine one of the seven levels of security of computer networks that are used in the United States Department of Defense.

Elements realization of software-algorithmic system optimization of technological processes based on adaptive methods

The paper examines a possibility to optimize complex technological processes of energy resources transfer and consumption. Notes the importance of unidentified and weakly formalized factors consideration, which influence quality of the described object classes functioning. Suggests use of fuzzy neural networks and controllers on their basis for implementation of technological processes optimization of energy resources transfer and consumption at the expense of its dynamic accuracy forecast improvement. Presents algorithmic realization, that later on configured in the MATLAB simulated environment. Quantitative and qualitative parameters, which characterize fuzzy neural controller’s usage quality, were received. Later, software environment elements are being described, which constructed on the basis of the suggested adaptive approach for complex technological processes optimization. Points out a possibility to conduct comparative study, and also simulation modeling implementation by way of presented solutions.

Прогнозування екологічного стану міської транспортної системи з використанням нейронних мереж

Мета: Метою цієї роботи є розробка ефективної моделі оцінювання екологічного стану міської транспортної системи з використанням загальних принципів нейронних мереж. Методи дослідження: Запропонована модель основана на роботі двох нейронних мереж з урахуванням впливу щільності транспортного потоку та рівня транзитного потенціалу на урбанізовані райони. Результати: На основі синтезу загальних принципів теорії нечітких множин та нейронних мереж розроблено модель оцінки рівня екологічного стану міської транспортної системи. Дано графічне представлення моделі. Високий ступінь достовірності прогнозу забезпечується навіть при низьких швидкостях навчання і високій динаміці змін статистичних даних в умовах транзиту транспортних потоків міста. Висновки: Використання нечітких нейронних мереж забезпечує можливість встановлювати повну відповідність між математичним представленням процедури нечіткого висновку і структурою міської транспортної системи. Запропонована модель дозволяє сформулювати чіткі екологічні орієнтири при прийнятті рішень в області транзитних перевезень з урахуванням інтересів підприємств, транспорту і населення міста, з подальшим перерозподілом транспортних потоків в тимчасовому і географічному просторі промислових районів міста.

A New Steganography Algorithm Using Hybrid Fuzzy Neural Networks

In recent years, image steganography has been one of the emerging research areas. As the field of information technology is advancing, the need of information security is increasing day by day. Steganography is a widely used communication method in today's scenario which involves sending secret information in appropriate carriers. Since it have an interesting property of concealing the message as well as the existence of the message, steganography is on its evolutionary path to unearth new platforms. As the field of steganalysis is growing exponentially, the need of developing strong steganographic algorithms is also growing. Since the use of steganography is spreading across various fields, the goal of increasing the embedding capacity, security and image quality is being major concerns. We propose a new image steganographic method which is based on random selection of pixels for secret data embedding and post processing the stego-image using Hybrid Fuzzy Neural Networks. The pixels where secret data is to be embedded is selected randomly using a pseudo random key. In the selected pixels the last 2 or 3 bits are used for hiding. The resultant degradation in the quality of stego-image is handled by an efficient pixel adjustment process with the use of fuzzy neural networks.. The experimental results reveal that this method can achieve an embedding capacity of 3 bits per byte with excellent stego-image quality and high imperceptibility.

Modeling of Self-Induced Vibrations that Occur During the Machining Process of Casting Patterns with the Use of The Fuzzy-Neural Networks Method

Abstract This article outlines a methodology of modeling self-induced vibrations that occur in the course of machining of metal objects, i.e. when shaping casting patterns on CNC machining centers. The modeling process presented here is based on an algorithm that makes use of local model fuzzy-neural networks. The algorithm falls back on the advantages of fuzzy systems with Takagi-Sugeno-Kanga (TSK) consequences and neural networks with auxiliary modules that help optimize and shorten the time needed to identify the best possible network structure. The modeling of self-induced vibrations allows analyzing how the vibrations come into being. This in turn makes it possible to develop effective ways of eliminating these vibrations and, ultimately, designing a practical control system that would dispose of the vibrations altogether.

The control application and simulation – particle swarm optimization exploration of control application for user intention toward mobile Mandarin learning system

This study introduces the use of fuzzy neural networks with intent to clearly visualize the advantages of the approach. Particle swarm optimization as the underlying optimization tool is discussed. Different models of fuzzy neural networks can be envisioned. It is worth stressing that the underlying concept is general enough and any other formalism of information granules could be considered as well. Over the past decades, the Chinese market has developed rapidly, especially on intelligent robot interaction and even tension leg platform. Consequently, Chinese learning has increased. E-learning has also become another learning mechanism and is already quite popular. In the study, optical character recognition software is developed in the android phone system. A questionnaire is designed as a base technology acceptance model, and, according to the dimensions, to make assumptions. Using a structural equation model analysis framework to analyze the recovery over the questionnaire, the conclusion of this study and the possible direction of future research are finally arrived at.

Research on fuzzy neural network algorithms for nonlinear network traffic predicting

This paper addresses the use of fuzzy neural networks (FNN) for predicting the nonlinear network traffic. Through training the fuzzy neural networks with momentum back-propagation algorithm (MOBP) and choosing the appropriate activation function of output node, the traffic series can be well predicted by these structures. From the effective forecasting results obtained, it can be concluded that fuzzy neural networks can be well applicable for the traffic series prediction. In addition, the performance of the FNN was particularly discussed and analyzed in terms of prediction ability compared with solely neural networks. The effectiveness of the proposed FNN is demonstrated through the simulation.

Use of fuzzy neural networks in modeling relationships of HPV infection with apoptotic and proliferation markers in potentially malignant oral lesions

To evaluate in oral leukoplakia the relationship between HPV infection and markers of apoptosis (bcl-2, survivin) and proliferation (PCNA), also conditionally to age, gender, smoking and drinking habits of patients, by means of Fuzzy neural networks (FNN) system 21 cases of oral leukopakia, clinically and histologically diagnosed, were examined for HPV DNA presence, bcl-2, survivin and PCNA expression. HPV DNA was investigated in exfoliated oral mucosa cells by nested PCR (nPCR: MY09-MY11/GP5-GP6), and the HPV genotype determined by direct DNA sequencing. All markers were investigated by means of standardised immunohistochemistry procedure. Data were analysed by chi-square test, crude OR and the 95% CI; in blindness, FNN was applied. HPV DNA was found in 8/21 OL (38.1%); survivin, PCNA, and tobacco smoking were associated in univariate analysis (p = 0.04) with HPV DNA status. HPV-18 was the most frequently detected genotype (6/8), followed by HPV-16 (2/8). FNN revealed that survivin and PCNA, both being expressed in all of OL HPV+ve, were associated with HPV infection. In conclusion, the FNN allowed to hypothesise a model of specific variables associated to HPV infection in OL. The relevance of survivin and PCNA suggest that they may be involved in HPV-mediated deregulation of epithelial maturation and, conversely, that HPV may have a role in the expression level of these two markers. FNN system seems to be an effective tool in the analysis of correlates of OL and HPV infection.

Expression of cell cycle markers and human papillomavirus infection in oral squamous cell carcinoma: Use of fuzzy neural networks

Our aim was to evaluate in oral squamous cell carcinoma (OSCC) the relationship between some cell cycle markers and HPV infection, conditionally to age, gender and certain habits of patients, and to assess the ability of fuzzy neural networks (FNNs) in building up an adequate predictive model based on logic inference rules. Eighteen cases of OSCC were examined by immunohistochemistry for MIB-1, PCNA and survivin expression; presence of HPV DNA was investigated in exfoliated oral mucosa cells by nested PCR (nPCR, MY09-MY11/GP5-GP6), and HPV genotype was determined by direct DNA sequencing. Data were analyzed by traditional statistics (TS) and FNNs. HPV DNA was found in 9/18 OSCCs (50.0 %) without any significant higher risk of HPV infection with respect to the sociodemographic variables considered (p > 0.2), apart from tobacco smoking, reported in 44.4% of OSCC HPV-positive vs. 100% HPV-negative subjects (p = 0.029). Regarding cell cycle markers, TS and FNN revealed that survivin was expressed significantly more in HPV-negative than in HPV-positive OSCC [root mean-square error (RMSE) = 5.89 x 10(-6), % predicted 100.0]; furthermore, smoking played a protective role for survivin expression in HPV-positive cases (OR = 0.019, 95%CI 0.001-0.723, RMSE = 0.20, % of prevision 94.4). FNN, although on a small sample size, allowed us to confirm data by TS and to hypothesize a different cell cycle pattern for HPV-positive vs. HPV-negative OSCC. In the latter cases, the relevance of apoptotic vs. proliferative markers suggested that they may be related to the different supposed outcome of HPV-negative OSCC and that HPV may have a protective role in the expression level of survivin, especially in tobacco smokers.