... The adaptive resonance networks discussed earlier implement a functionally similar rule using the difference between the weight and the element of the input to the node. Note if all weights are equinormal, then the two different activations rules are functionally the same (Kosko, 1992) pp. 147 and competitive-type activation rules (based on differences between input and stored experience) reduce to the correlation measure. ...
... See also the treatment of (Kosko, 1992). This abstraction is referred to by Shepherd (1994) as the simple synaptic model of adaptation corresponding to Hebb's postulate of learning (Hebb, 1949). ...
... and then re-writing using the differential form of the Hebbian learning rule (see (Kosko, 1992) for further discussion): ...
p>This thesis presents a theory of computational agency. Computational agent theory differs from 'classical' artificial intelligence by committing to the view that a computational artifact is situated, and that its rationality is limited by the constraints of this 'situatedness'. Contempo- rary literature is surveyed and models of situated computational agency placed in their philo- sophical contexts. From this, a critical reconstruction of the notion of an agent is given from a phenomenological perspective. It is proposed that everyday routines of activity underpins agency and computational implementations of this substrate can take the form of connectionist networks. The proposal is tested in technical practice on two domains; agents for multimedia content- based navigation/retrieval, and a simulated environment which explores the key properties of the proposed phenomenological agent theory. Recent proposals for goal-directed behaviour in connectionist systems (largely from the cognitive and behavioural neurosciences) are critically evaluated, and integrated into an agent architecture. This results in an architecture utilising suit- ably controlled reinforcement learning. The architecture implemented is then evaluated against the agent theory, and examples of 'routine behaviour' analysed in stationary and non-stationary environments. Semiotic analyses are then proposed as an alternative theory of representation, as they are compatible with, and simultaneously possess explanatory power at a level beneath, the usual sentential/propositional level. The thesis contributes a phenomenological theory of agency and gives examples of its influence on technical practice. Outlines of connectionist architectures are presented, imple- mented, and evaluated with respect to the agent theory proposed.</p
... Рассмотрим вначале простейшую составляющую любой когнитивной карты: два элемента 1 Ax выражается, как правило, числом, которое и является весом 12 a . Это соотношение причинного следования имеет и различные обоснования с точки зрения формальной логики, но останавливаться здесь на этом не будем. ...
... Динамическая сторона НКК состоит в том, что данный подход в настоящее время развился в отдельную методологию моделирования, которая рассматривает набор концептов (которые могут выражать собой не только понятия, но и любые характеристики системы, в том числе и физические) и взвешенные связи между ними как дискретную динамическую систему. В некотором смысле такая система подобна системе ОДУ со специфическим поведением траекторий (точки равновесия, предельные циклы, хаос и т.п.), см., например, раннюю работу [12] и более современные сборники статей [13 -15]. В последнее время усилился интерес к более сложным НКК, где как концепты, так и отношения между концептами сохраняют нечеткий характер на всех этапах моделирования. ...
... Пусть пока, для простоты, это будут некоторые числовые значения. Тогда отношение влияния в самом простом виде можно записать как 2 A -просто коэффициент связи будет равен 12 1/ a , то есть связь двух концептов вне системы не имеет смысла. Если имеется несколько концептов, то возникает система, которую будут определять структура и веса связей, а также метод вычисления значений концептов. ...
A broader view of the technology of fuzzy cognitive maps is described, in which the cognitive map is considered as a carrier of computational procedures. This approach can be described as a generalized system dynamics. This interpretation makes it easier to obtain theoretical results that can characterize the behavior of complex systems. In particular, in the case of simple computational procedures, the relationship between the degree of influence of factors and the structure of the system, namely, the presence of connecting paths and cycles in the corresponding digraph, is clarified.
... FCMs form a field of intelligent modeling and computing that has gained constantly increasing research interest in the last ten years. FCMs constitute graphical models in the form of directed graphs consisting of two basic elements: Nodes, which correspond to Cognitive Concepts bearing different states of activation depending on the knowledge they represent; Arrows denoting the causal effects that source nodes exercise on the receiving concept expressed through weights [33,34]. It can be considered as an integration of multifold subjects, including neural network, graph theory and fuzzy logic. ...
... FCM's consists of nodes, called "concepts" and weighted interconnections among them, called "weights". Concepts denote entities, states, variables, or characteristics of the system under investigation [34]. Using mathematical terms, a FCM can be defined using a 4-tuple ðC; W; A; fÞ where C ¼ C 1 ; C 2 ; . . ...
Fuzzy Cognitive Maps (FCMs) have gained popularity within the scientific community due to their capabilities in modeling and decision making for complex problems. In order to examine and estimate the FCM inference, we develop and the ‘fcm’ open-source package R package. Specifically, the fcm.infer function provides six inference rules and four threshold functions. This open-source ‘fcm’ package is available in CRAN and provides the opportunity for everyone to run different scenarios in their weighted matrices. Dedicated examples and visualizations are provided to show the proposed open-source FCM package for business modelling and decision making.
... When the overlapping exits between two membership functions then they should not have the same point of maximum truth and the sum of truth for any point within the overlap must be less than or equal to one. Regarding setting the completeness level, some authors [5] suggest a minimum completeness level =0.25 and others [6] suggest an average completeness level=0.50. However, it is dependent up on the set of fuzzy rules and application domain. ...
This paper describes a suitable fuzzy membership function and input/output parameter for diagnosis of Rheumatic Fever (RF) in Nepal. The current case in Nepal is that computerized health informatics systems and appropriate data sets of RF are not available. Also, some signs and symptoms of RF do not reflect the measurement numerical values. These signs and symptoms will expresses in linguistic variables based on the doctor’s belief. Therefore, it is quite hard to determine the membership functions and input/output parameter. We purposed manual adjustable methods to determine the input/output parameters and membership function based on the predefine set of fuzzy rules. We tested different membership functions and changed the output parameter values until the suitable result is not achieved. We applied this method to diagnoses of arthritis pain and evaluated a system in Matlab Fuzzy toolbox. In this paper, we focus on the evaluation the different membership functions with different input/output parameters and present the results.
... Teori himpunan fuzzy ini didasarkan pada logika fuzzy (Kosko, B. 1992). Terdapat nilai logika antara 0 dan 1 yang menyatakan tingkat kebenaran. ...
E-Book-01: Fuzzy and Artificial Neural Networks Programming for Intelligent Control Systems (Indonesian version)
How to program Fuzzy and Artificial Neural Networks Using Python and implemented in Intelligent Control System
... The authors attached fuzzy set as a triangular fuzzy number (Kosko, 1992) for each rating. This unique way of describing the criteria made it possible to evaluate the variants in a way that is common in education. ...
The choice of agricultural crops for sowing and the planning of yields and economic results is realized in conditions of uncertainty and risk. The factors that contribute most to the uncertainty in achieving yields in agricultural production can be quantified using the method of fuzzyfication. The method based on the unique assessment of criteria represents an innovative approach to solving uncertainty in agricultural production. A key feature of this method is its ability to treat criteria as fuzzy scores and to allow their aggregation to make a final yield planning decision. In the paper, 15 main criteria were chosen that influence the planning of yield when sowing sage, nettle and rye. Based on economic indicators, the economic analysis does not numerically describe the impact of uncertainty, which in agricultural production can have an inestimable importance on realized yields and incomes. The method used is a suitable tool for analysis and planning in agriculture, it enables effective treatment of uncertainty and competing criteria, providing farmers with a reliable basis for making decisions about yield planning.
... A MLFF-ANN; is trained through error back propagation (EBP) technique for predicting the tool wear. The EBP is supervised learning technique based on generalized delta rule [13] which requires input and output sets referred as training patterns. ...
A FSW is solid-state joining process that combines two angled pieces of metal without melting the metal. The process generates heat by friction among angled metal pieces and rotating tool, producing heat that is then used to fuse the pieces together. This leads to a softened area around the FSW tool. In the present work the effects of several factors of the FSW welding process i.e. rotation speed, welding speed and Shoulder Diameter, have been investigated to analyze the output characteristic. The output parameters were forecasted using multi-layer feed forward ANN (MLFF-ANN) and analyzed using analysis of variance. Confirmation experimental runs were also conducted in order to validate the results.
... ; SðA i ; B m B Þg and B j ¼ fSðB j ; A 1 Þ; . . . ; SðB j ; A m A Þg :S(A i , B j ) denotes the subsethood[61,62] of module A i in module B j , and S(B j , A i ) denotes the subsethood of module B j in module A i :SðA i ; B j Þ ¼ jA i \ B j j jA i j and SðB j ; A i Þ ¼ jA i \ B j j jB j j : Typically S(A i , B j ) 6 ¼ S(B j , A i). Thus, h A!B 2 [0, 1] captures how much the modules of partition fA 1 � � � A m A g are on average dispersed into the modules of partition fB 1 � � � B m B g, and h B!A 2 [0, 1] the other way around. The smaller the values are, the more the community structure of one network is preserved in the other. ...
The structure of social networks strongly affects how different phenomena spread in human society, from the transmission of information to the propagation of contagious diseases. It is well-known that heterogeneous connectivity strongly favors spread, but a precise characterization of the redundancy present in social networks and its effect on the robustness of transmission is still lacking. This gap is addressed by the metric backbone, a weight- and connectivity-preserving subgraph that is sufficient to compute all shortest paths of weighted graphs. This subgraph is obtained via algebraically-principled axioms and does not require statistical sampling based on null-models. We show that the metric backbones of nine contact networks obtained from proximity sensors in a variety of social contexts are generally very small, 49% of the original graph for one and ranging from about 6% to 20% for the others. This reflects a surprising amount of redundancy and reveals that shortest paths on these networks are very robust to random attacks and failures. We also show that the metric backbone preserves the full distribution of shortest paths of the original contact networks-which must include the shortest inter- and intra-community distances that define any community structure-and is a primary subgraph for epidemic transmission based on pure diffusion processes. This suggests that the organization of social contact networks is based on large amounts of shortest-path redundancy which shapes epidemic spread in human populations. Thus, the metric backbone is an important subgraph with regard to epidemic spread, the robustness of social networks, and any communication dynamics that depend on complex network shortest paths.
... В його моделі визначається співвідношення ракет, яке із великою ймовірністю не призведе до війни. Визначається оптимальна стратегія для сторони, що нападає: завдати по супротивнику рішучого тотального удару, або удару контрольованого 4 . ...
Using the example of the modern military conflict between the Russian Federation and Ukraine, the article reveals the essence of the concept of "asymmetric conflict" as an ideal type, which allows building a network of concepts for the purpose of further studying the phenomenon of hostilities, as well as building forecasts of the development of the military-political situation.
The conceptual provisions of the organization of asymmetric countermeasures against the aggression of a "more powerful" state in military terms by the "victim state" regarding the protection of its fundamental vital national interests are substantiated. The specific focus is the application of mathematical modeling in the framework of the military confrontation between the Russian Federation and Ukraine. The possibilities of introducing the principles of mathematical modeling into the methodology of political science are analyzed and outlined.
The article puts forward a hypothesis regarding which the range of application of fuzzy models and methods is quite suitable for the subject of the proposed study, given that the unpredictability of the result should be attributed to the peculiarities of asymmetric conflicts in the event of a clear discrepancy in the power capabilities of the opposing sides ("indirect" tactics, non-standard military actions applied by the weak side, as well as the inability of the strong side to defend its positions and finally suppress the weaker opponent).
On the basis of expert evaluations, using the example of the Russian-Ukrainian conflict, the use of a fuzzy cognitive map (FCM) is shown in order to model the causal relationships that are identified between the concepts in the conflict model.
Taking into account the potential capabilities of the warring parties, the level of the possibility of a nuclear threat from the Russian Federation was theoretically predicted and the ranking of factors that could affect Ukraine’s victory in the military conflict was carried out.
... A MLFF-ANN; is trained through error back propagation (EBP) technique for predicting the tool wear. The EBP is supervised learning technique based on generalized delta rule [16] which requires input and output sets referred as training patterns. Appropriate weights to predict the desired output is done by MLFF-ANN. ...
This electronic When precision is required, WEDM is a conductive material machining technology that is widely used. The rough cutting operation in WEDM is regarded tough because precision work demands the use of numerous machining performance measurements, such as metal removal rate (MRR), surface finish (SF), and cutting width (kerfs).According toResponse Surface Methodology, discharge current, pulse frequency, wire-speed, wire tension, and dielectric flow are all key machining characteristics that influence performance metrics (RSM). Each performance metric's best set of parameters has been discovered to be different. This study uses nonlinear regression analysis to stabilize the relationship between control variables and response variables like MRR, SF, and kerf, resulting in a valid mathematical model. Little research has been done to establish the ideal quantity of machining parameters for the best machiningquality in tough machining materials such as hot die steel D4 according to the literature review. Hot die steel D4 is widely utilized in applications such as hot work forging, extrusion, punching tools, mandrels, mechanical press forging dies, plastic molds and die casting dies, aviation landing gears, helicopter rotor blades, and shafts, and others. Wire electrical discharge machining is challenging to obtain consistent quality because the process parameters are difficult to maintain. The most challenging problems for researchers and engineers to address are those listed above. Manufacturers strive to find control elements to increase machining quality based on their operational experiences, manuals,or failed attempts. Electronic Process Tools Limited's wire cut EDM was used to machine D4 hot die steel specifically for this application (Elektra Spirit cut 734).
... В рамках настоящей публикации мы остановимся на моделях так называемых нечетких когнитивных карт, то есть на системе факторов, организованных в направленный граф, вершины и ребра которого снабжены весами той или иной природы. Эта тематика очень широко представлена в мировой литературе, получить представление о ней можно, например, по «базовым» публикациям основателя данного направления Б. Коско [7][8][9]. Здесь мы будем использовать модификацию технологии когнитивных карт, которая проводит определенную параллель между когнитивными картами и нейросетями, см. [10,11], где также представлен небольшой обзор текущих направлений развития технологий нечетких когнитивных карт. ...
The preprint is an example of the practical implementation of a modified modeling methodology using fuzzy cognitive maps, which was presented by the author earlier in the publication in "M.V. Keldysh IPM Preprints". A variant of the model of the development of the RF region in conditions of high uncertainty of the external environment and with a lack of initial data is described. This situation is typical, for example, for the moments of the outbreak of epidemics or conflict confrontation.
... There is an interesting geometric analog for illustrating the idea of set membership (Kosko, 1992). Heretofore we have described a fuzzy set A defined on a universe X. ...
Fuzzy control has increased tremendous interest in applications over the past few years and also among control equipment. The present book titled “Fuzzy Logic Models and Fuzzy Control: An Introduction” has been written to meet these inspirations. It consists of total nine chapters: First three chapters are related to fuzzy set theory, fuzzy logic, fuzzy
systems and models, while next six chapters deal with design and analysis methodology
of fuzzy control. However, it should be clear that such a universal theory does not exist for conventional control engineering either, so we have to proceed from a few isolated spots
where we already know exactly how to design a fuzzy control algorithm to clusters of problems and related design methodologies.
... • The recurrence relation with the threshold function (Kosko, 1992) does not provide sensitivity of the output function at the levels of input variables within the interval [0,1] (this will be shown in Section 3.5). • A reliability model in the form of an FCM with expert values of arc weights without adjustment stage does not guarantee the proximity of simulation results and experimental data. ...
This article offers a method for analyzing the reliability of a man–machine system (MMS) and ranking of influencing factors based on a fuzzy cognitive map (FCM). The ranking of influencing factors is analogous to the ranking of system elements the probabilistic theory of reliability. To approximate the dependence of “influencing factors—reliability,” the relationship of variable increments is used, which ensures the sensitivity of the reliability level to variations in the levels of influencing factors. The novelty of the method lies in the fact that the expert values of the weights of the FCM graph edges (arcs) are adjusted based on the results of observations using a genetic algorithm. The algorithm's chromosomes are generated from the intervals of acceptable values of edge weights, and the selection criterion is the sum of squares of deviations of the reliability simulation results from observations. The method is illustrated by the example of a multifactor analysis of the reliability of the “driver–car–road” system. It is shown that the FCM adjustment reduces the discrepancy between the reliability forecast and observations almost in half. Possible applications of the method can be complex systems with vaguely defined structures whose reliability depends very much on interrelated factors measured expertly.
... In particular, the fusion of fuzzy reasoning (FR), artificial intelligence (AI) and neural networks (NN) has attracted more attention. Some scholars call it the FAN system [26] and believe that the FAN system is a very practical information processing system with a computer as the main platform [27]. ...
The fuzzy logic reasoning based on the “If... then...” rule is not the inaccurate reasoning of AI against ambiguity because fuzzy reasoning is antilogical. In order to solve this problem, a redundancy theory for discriminative weight filtering containing six theorems and one M(1,2,3) model was proposed and the approximate reasoning process was shown, the system logic of AI handling ambiguity as an extension of the classical logic system was proposed. The system is a generalized dynamic logic system characterized by machine learning, which is the practical-application logic system of AI, and can effectively deal with practical problems including conflict, noise, emergencies and various unknown uncertainties. It is characterized by combining approximate reasoning and computing for specific data conversion through machine learning. Its core is data and calculations and the condition is “sufficient” high-quality training data. The innovation is that we proposed a discriminative weight filtering redundancy theory and designed a computable approximate reasoning logic system that combines approximate reasoning and calculation through machine learning to convert specific data. It is a general logic system for AI to deal with uncertainty. The study has significance in theory and practice for AI and logical reasoning research.
... Indeed, the subject of AISs is fascinating and productive. For further information, the reader is referred to [48], [49], [50], [51], [52]. ...
This paper presents an overview of the methodologies and applications of artificially intelligent systems (AIS) in different engineering disciplines with the objective of unifying the basic information and outlining the main features. These are knowledge-based systems (KBS), artificial neural networks (ANN), and fuzzy logic and systems (FLS). To illustrate the concepts, merits, and demerits, a typical application is given from each methodology. The relationship between ANN and FLS is emphasized. Two recent developments are finally presented: one is intelligent and autonomous systems (IAS) with particular emphasis on intelligent vehicle and highway systems, and the other is the very large scale integration (VLSI) systems design, verification, and testing.
... This proposition forms the basis of the Differential Hebbian Learning (DHL). Kosko (1992) discusses the use of DHL as a form of unsupervised learning for FCMs. DHL can simplify the construction of FCMs by allowing the expert to enter approximate values (or even just the signs) for causal link strengths. ...
1 paper - 1. Lyazzat Atymtayeva, Gulfarida Tulemissova, Serik Nurmyshev. An Intelligent Approach and Data Management in Active Security Auditing Processes for Web Based Applications, Proceedings of the Seventh International Symposium on Business Modeling and Software Design, Barcelona, Spain , 3-5 July 2017, p.136.
2paper - 2. Lyazzat Atymtayeva, Gulfarida Tulemissova, Serik Nurmyshev and Ardakbek Kungaliyev, Some Issues in the Re-Engineering of Business Processes and Models by Using Intelligent Security Tools, Proceedings of the Seventh International Symposium on Business Modeling and Software Design, Barcelona, Spain , 3-5 July 2017, p. 199
... In [1], Axelrod used a directed graph to describe the connections between the political elites. This modelling technique was extended by Kosko [23,24], who introduced Fuzzy Cognitive Maps (FCMs), by representing the strength of the causal connections using values from the ½À1; 1 interval. The nodes of the graph represent the main subsystems, or system variables; while the weighted, directed edges express the causal knowledge [39]. ...
Complex systems can be effectively modelled by fuzzy cognitive maps. Fuzzy cognitive maps (FCMs) are network-based models, where the connections in the network represent causal relations. The conclusion about the system is based on the limit of the iteratively applied updating process. This iteration may or may not reach an equilibrium state (fixed point). Moreover, if the model is globally asymptotically stable, then this fixed point is unique and the iteration converges to this point from every initial state. There are some FCM models, where global stability is the required property, but in many FCM applications, the preferred scenario is not global stability, but multiple fixed points. Global stability bounds are useful in both cases: they may give a hint about which parameter set should be preferred or avoided. In this article, we present novel conditions for the global asymptotical stability of FCMs, i.e. conditions under which the iteration leads to the same point from every initial vector. Furthermore, we show that the results presented here outperform the results known from the current literature.
... Today, fuzzy set theory is widely used in expert systems, neural networks and artificial intelligence systems, and the world has a sufficient number of scientific papers on this topic. The theoretical foundations of fuzzy set theory are laid in the works of B. Cosco (1986;1991), the author of the FAT theorem (Fuzzy Approximation Theorem). According to this theorem, any classical mathematics can be approximated by means of fuzzy logic. ...
The rapid development of computer technology has led to the use of fuzzy set theory in the medical, financial, economic, commercial and other fields as one of the basic components of artificial intelligence. This is due to its universal mechanism designed to analyze research in the field of humanities research. The mathematical apparatus of fuzzy set theory allows them to be performed with unformalized data, and the development and improvement of information and communication technologies make it possible to automate this process. Unformalized, abstract, "blurred" statistics, which are difficult to analyze, are also common in pedagogy. But in pedagogical practice, fuzzy logic has not been widely used. The article proves the importance and expediency of teaching students, future teachers of computer science, skills in the application of fuzzy set theory. The ability to use the mechanisms of fuzzy logic in applied programs will allow future teachers in their further pedagogical activities to conduct multi-criteria analysis of various characteristics of their students, analysis of pedagogical methods, comprehensive assessment of competencies and more. The article presents the experience of teaching fuzzy set theory, the logic of teaching and its sequence, as well as the results of such training at a pedagogical university. The necessity of the step-by-step study of fuzzy set theory is proved - from acquaintance with its basic concepts, giving examples of its application in expert systems, neural networks and artificial intelligence systems to independent construction of fuzzy knowledge representation model, development of linguistic variables and use of spreadsheet or specialized programs. The results of the experimental introduction of the topic "Fuzzy model of knowledge representation" in a training course of computer disciplines are shown. Examination of learning outcomes reveals a positive attitude of the students toward mastering the skills of using fuzzy set theory and willingness to apply it in their further pedagogical activities.
... Fuzzy Cognitive Maps (FCMs) introduced by Kosko (1986) extend the idea of Cognitive Maps by allowing the concepts to be represented linguistically with an associated fuzzy set. FCMs are fuzzy signed digraph with feedback (Kosko, 1986(Kosko, , 1988. It represents causal relationship between concepts. ...
The employee-employer relationship is an intricate one. In an industry, the employers expect to achieve performances in quality and production in order to earn profit, on the other side employees need good pay and all possible allowances and best advantages than any other industry. Our main objective of this paper is to analyze the relationship between employee and employer in workplace and discussed how to maintain a strong employee and employer relationship which can produce the ultimate success of an organization using Induced Fuzzy bi-model called Induced Fuzzy Cognitive Relational Maps (IFCRMs). IFCRMs are a directed special fuzzy digraph modelling approach based on expert's opinion. This is a non statistical approach to study the problems with imprecise information.
... Các nghiên cứu về nhận biết và phân loại tập mờ đã được nhiều nhà nghiên cứu thực hiện, điển hình là các nghiên cứu nhằm kết hợp giữa logic mờ và mạng nơron để phát triển các hệ thống thông minh [2,3,4]. Điều này nhằm kết hợp sức mạnh của lý luận mờ trong xử lý thông tin không chắc chắn với khả năng học của mạng nơron [5,6,7,8]. Lin [9] dựa trên các ý tưởng của điều khiển logic mờ, cấu trúc mạng nơron và khả năng học của mạng nơron để đưa ra mô hình điều khiển logic mờ và hệ thống quyết định. Archer [10], Mukaidono [11] sử dụng tập mờ trong mạng nơron phân lớp. ...
1 Trường Đại học Công nghệ Thông tin và Truyền thông Thái Nguyên 2 Trường Cao đẳng Công nghiệp Thái Nguyên e-Mail: 1 anhtucntt@gmail.com, 2 vmc802@gmail.com Tóm tắt Bài báo này trình bày một mạng nơron được cải tiến từ mô hình mạng nơron phân cụm mờ min-max (FMNN). Mô hình cải tiến được gọi là GFMNN nhằm giải quyết hai hạn chế căn bản của FMNN. Thứ nhất, GFMNN có cấu trúc tự động tăng trưởng số lượng nơron trong quá trình huấn luyện tùy theo đặc tính riêng của mỗi tập dữ liệu. Thứ hai, GFMNN cho phép xử lý trực tiếp các dữ liệu thực mà không cần chuẩn hóa trước. Các thực nghiệm đã được chúng tôi tiến hành trên hai tập dữ liệu PID và Iris để so sánh GFMNN với các phương pháp khác đã được công bố. Từ khóa: min-max mờ, mạng nơron, phân cụm, phân lớp, tăng trưởng. Abstract: This paper presents an impoved fuzzy min-max neural network (FMNN). The improved model is called Growing FMNN (GFMNN) which aims to overcome the two drawbacks of FMNN. Firstly, GFMNN has growing structure which automatically increases the number of neurons during training depending on specific characteristics of each dataset. Secondly, GFMNN allows directly processing real data without standardizing. The experiments were conducted on two data sets IRIS and PID to compare GFMNN with other methods which have been published previously. 1. Phần mở đầu Lý thuyết tập mờ được Zadeh đưa ra vào năm 1965 [1]. Các nghiên cứu về nhận biết và phân loại tập mờ đã được nhiều nhà nghiên cứu thực hiện, điển hình là các nghiên cứu nhằm kết hợp giữa logic mờ và mạng nơron để phát triển các hệ thống thông minh [2,3,4]. Điều này nhằm kết hợp sức mạnh của lý luận mờ trong xử lý thông tin không chắc chắn với khả năng học của mạng nơron [5,6,7,8]. Lin [9] dựa trên các ý tưởng của điều khiển logic mờ, cấu trúc mạng nơron và khả năng học của mạng nơron để đưa ra mô hình điều khiển logic mờ và hệ thống quyết định. Archer [10], Mukaidono [11] sử dụng tập mờ trong mạng nơron phân lớp. Dựa trên những ưu điểm của việc kết hợp logic mờ và mạng nơron, Simpson đã đề xuất một mô hình mạng nơron phân cụm mờ min-max, gọi là FMNN [14,15], cho phép kết hợp mạng nơron và lý thuyết min-max mờ để giải quyết bài toán phân lớp và phân cụm. FMNN dựa trên sự tổng hợp của các hyperbox mờ [16] để xác định và giới hạn các không gian con trong không gian mẫu n-chiều. Mỗi không gian con được xác định bằng hai điểm min và max được gọi là một hyperbox. Nói cách khác, FMNN phân cụm bằng cách tạo ra các hyperbox, mỗi hyperbox đại diện cho một cụm sẽ tương ứng là một nơron trong lớp đầu ra. Số lượng nơron của lớp đầu ra được ước lượng cố định trước sao cho lớn hơn số cụm có thể hình thành. Thuật toán học của FMNN nhằm mục tiêu mở rộng và co lại các hyperbox thông qua việc điều chỉnh trọng số của mạng, nếu quá trình mở rộng tạo ra sự chồng lấn giữa các hyperbox thì thực hiện quá trình co lại để khử chồng lấn. Đến nay, có một số nghiên cứu đề xuất cải tiến hiệu suất thuật toán học của FMNN. Rizzi đã cải tiến FMNN của Simpson bằng cách áp dụng các kỹ thuật phân loại thích ứng [17], cắt tỉa [18], khái quát PARC (Pruning Adaptive Resolution Classifier) [19] sử dụng kỹ thuật cài đặt đệ quy. Tuy nhiên, các giải pháp này có chi phí tính toán lớn, do hạn chế của việc sử dụng đệ quy [18]. Meneganti và nhóm nghiên cứu [20,21] đề xuất thuật toán học bằng cách bắt đầu với số các hyperbox bằng đúng với số hyperbox đầu ra. Thuật toán này phụ thuộc vào thứ tự trình bày các dữ liệu và các kích thước của hyperbox. Các giải pháp học của cả Rizzi và Meneganti là "offline" do phải duyệt lại mẫu huấn nhiều lần để xác định các cụm. Trong các thuật toán này, đều cần dự báo trước số lượng các hyperbox cần được tạo ra. Việc dự báo trước chính xác các hyperbox là đặc biệt khó đối với loại dữ liệu có độ nhiễu cao, số cụm lớn. Trong bài báo này, chúng tôi đề xuất một mô hình mạng nơron mờ truyền thẳng có khả năng tự tăng trưởng trong quá trình huấn luyện. Mô hình này được cải tiến từ mô hình của Simpson. Khả năng tự tăng trưởng trong quá
Liquefaction is a phenomenon whereby a saturated or partially saturated soil substantially loses strength and stiffness due to combine action of ground acceleration and increase in pore water pressure. This phenomenon is common in sandy soil. Many various methods have been used to estimate the liquefaction potential of a region based on available data, like SPT-N value method, because it's simplicity and requirement of less variables. This study is aimed to adopt SPT-N based technique for liquefaction assessment of soil for rivers coast regions of Allahabad (India) city. For the purpose of finding liquefaction potential Modified Seed's method was first used to develop liquefaction database of different sites. Further corrected SPT-N based values and liquefaction database through Modified Seed's method were used to develop Neuro-Fuzzy models. To develop Neuro-Fuzzy models bore log data were used. For the purpose of finding liquefaction potential Modified Seed's method was first used to develop liquefaction database of different sites. Further corrected SPT-N based values and liquefaction database through Modified Seed's method were used to develop Neuro-Fuzzy models Five pertinent inputs were used to model liquefaction. These variables were depth, corrected SPT-N value, bulk density, natural moisture content and percentage of particles finer than 0.075mm.
The current situation in economic activity is notable for exponential increase in accessible e-information and serious interest in its use in order to get competitive advantages. The article studies influence of information published in mass media on the essential social and economic indicators, which show key concepts of the system of strategic management in Russia. Methodological basis of the research was formed by theories of cognitive, topical modeling and regressive analysis. In the investigation the author used methods of topical modeling, machine education and statistical analysis of data. The author put forward the procedure of automated plotting of the cause-and-effect diagram based on qualitative and quantitative data. The system of causal links between key concepts of the system gave a chance to build forecast models of high accuracy. Findings of the research showed that topics being highlighted in mass media can influence social and economic indicators. Unfortunately, accidental events can make mathematic models, relying on system inertia, inadequate. In theoretical aspect the article proposes the procedure of automated building of the cause-and-effect diagram based on heterogeneous data, which can eliminate the problem of subjectivity of expert estimations in plotting cognitive maps. In applied aspect models of forecasting the most important social and economic indicators based on mass media publications were worked out that lean against cause-and-effect diagram and can support well-grounded managerial decisions and in case of necessity can affect the situation.
Advances on Mathematical Modeling and Optimization with Its Applications discusses optimization, equality, and inequality constraints and their application in the versatile optimizing domain. It further covers non-linear optimization methods such as global optimization, and gradient-based non-linear optimization, and their applications.
Discusses important topics including multi-component differential equations, geometric partial differential equations, and computational neural systems
Covers linear integer programming and network design problems, along with an application of the mixed integer problems
Discusses constrained and unconstrained optimization, equality, and inequality constraints, and their application in the versatile optimizing domain
Elucidates the application of statistical models, probability models, and transfer learning concepts
Showcases the importance of multi-attribute decision modeling in the domain of image processing and soft computing
The text is primarily for senior undergraduate and graduate students, and academic researchers in the fields of mathematics, statistics, and computer science.
The Russian–Ukrainian conflict is considered as a dynamic system, whose variables are factors affecting the losses of the Russian army and the threat of the use of nuclear weapons. A fuzzy cognitive map (FCM) is used for modeling, that is, a directed graph whose vertices are model variables, and the weights of arcs are the degrees of positive and negative influences of variables on each other. The following factors influencing the losses of the Russian army and the threat of a nuclear strike were selected: resistance of the Ukrainian army, support of Ukraine with weapons, economic sanctions against Russia, opposition to the Russian government and its self-preservation instinct. The degrees of the influence of factors on each other and on the possibility of using nuclear weapons are evaluated by experts using fuzzy terms, which correspond to numeric values. To adjust the FCM, a genetic algorithm is used to select the degrees of influence of factors that minimize the discrepancy between the simulation results and expert estimations. The obtained FCM is used for scenario modeling of the conflict according to the “what if” scheme and ranking of factors according to their degree of influence on the level of nuclear threat.
Load forecasting plays an essential role in the power system energy management system. Important factors that are crucial for short-term load forecasting (STLF) are unit commitment and economic allotment of the generation preservation schedule. At present, many techniques have been used for load forecasting, but Artificial Intelligence-based techniques such as Fuzzy Logic (FL) and Artificial Neural Networks (ANNs) gave better efficiency as compared to conventional techniques such as Regression and Time Series approaches. This chapter solves the STLF problem using FL, ANNs, and Jaya optimization algorithm methods and compares their errors to find out which one is more accurate for load forecasting. MATLAB, Simulink, the Fuzzy Logic Toolbox, and the ANN Toolbox are the programs utilized for this work.KeywordsLoad forecastingFuzzy logicArtificial neural networkEvolutionary algorithmJaya algorithm
This text was written as part of the mini-course presented at the Information Technology Update Journey at the 1996 Brazilian Computer Society Congress.
The objective of this mini-course is to provide a comprehensive view of the area of Artificial Neural Networks, presenting from the philosophical issues involved in the approach, through the theoretical aspects and ways of modeling the problems, to practical applications. The course
aims to introduce and motivate the student to pursue advanced studies in the area, presenting Neural Networks as a robust and efficient computational approach that can be used
in solving a variety of real-world problems.
In the initial part of this text, a review of the work carried out in Neural Networks is presented, how it emerged and how it developed historically. In the next step, some philosophical questions are presented with considerations about the need for the development of intelligent machines and the motivation for the use of neural networks. Next, we move on to the Modeling module, where a brief description of the area is presented, emphasizing the basic assumptions. The main models of neural networks and existing learning paradigms are described. The mini-course continues with theoretical considerations about the Computational Capacity of neural architectures and what can be learned by a neural network. In the final step, current examples of the countless number of practical problems where neural networks have found wide application and success are described, and their use in one of the areas of greatest application, Pattern Recognition, is illustrated. More particularly problems in Computer Vision, Image and Signal Processing are mentioned.
Este texto foi escrito como parte do minicurso apresentado na Jornada de Atualização em Informática no Congresso da Sociedade Brasileira de Computação de 1996.
O objetivo deste minicurso é proporcionar uma visão abrangente da área de Redes Neurais Artificiais, apresentando desde as questões filosóficas envolvidas na abordagem, passando pelos aspectos teóricos e as formas de modelagem dos problemas, até as aplicações práticas. O curso
visa introduzir e motivar o aluno a prosseguir estudos avançados na área, apresentando Redes Neurais como uma abordagem computacional robusta e eficiente e que pode ser empregada
na solução de diversos problemas do mundo real.
Na parte inicial deste texto, é apresentada uma revisão dos trabalhos realizados em Redes Neurais, como surgiu e como se desenvolveu historicamente. Na etapa seguinte são apresentadas algumas questões filosóficas com considerações sobre a necessidade do desenvolvimento de máquinas inteligentes e a motivação para a utilização de redes neurais. Em seguida, passa-se ao módulo de Modelagem onde é apresentada a uma breve descrição da área enfatizando os pressupostos básicos. São descritos os principais modelos de redes neurais e paradigmas de aprendizagem existentes. O minicurso segue realizando as considerações teóricas sobre a Capacidade Computacional das arquiteturas neurais e sobre o que pode ser aprendido por uma rede neural. Na etapa final, exemplos atuais da inúmera quantidade de problemas práticos onde redes neurais têm encontrado ampla aplicação e sucesso s\~ao descritos, e passa-se a ilustrar a sua utilização numa das áreas de maior aplicação, o Reconhecimento de Padrões. Mais particularmente são mencionados problemas em Visão Computacional, Processamento de Imagens e de Sinais.
This paper affords an impression device implementation found out for the electrical force device of a pump station supported analysis, this country of the control approach for the electrical force device of a pump station, and present challenges. The operation’s findings display that this type of control plan is feasible , which with the help of using incorporating computing , it'll successfully keep guide diligence while conjointly sleuthing pump station failures, thence growing the voltage pleasant of the power provide device. A nonlinear proportional operation approaches the transmission sign of the pumping station’s electric force device, and additionally the sign enters the nonlinear perform reckoner of the neural community controller, anyplace the neural community’s nonlinear mapping capacity is hired to understand computing control via the perform. Design questioning gives a framework that allows us. to spice up the proper query on the proper time while constructing partner AI solution. Moreover, it promotes a mind-set that accepts uncertainty, an massive component for varied superior AI comes Keywords: Electric force, Pump station, AI, Power supply system
The formation of patterns is one of the main stages in logical data analysis. Fuzzy approaches to pattern generation in logical analysis of data allow the pattern to cover not only objects of the target class, but also a certain proportion of objects of the opposite class. In this case, pattern search is an optimization problem with the maximum coverage of the target class as an objective function, and some allowed coverage of the opposite class as a constraint. We propose a more flexible and symmetric optimization model which does not impose a strict restriction on the pattern coverage of the opposite class observations. Instead, our model converts such a restriction (purity restriction) into an additional criterion. Both, coverage of the target class and the opposite class are two objective functions of the optimization problem. The search for a balance of these criteria is the essence of the proposed optimization method. We propose a modified evolutionary algorithm based on the Non-dominated Sorting Genetic Algorithm-II (NSGA-II) to solve this problem. The new algorithm uses pattern formation as an approximation of the Pareto set and considers the solution’s representation in logical analysis of data and the informativeness of patterns. We have tested our approach on two applied medical problems of classification under conditions of sample asymmetry: one class significantly dominated the other. The classification results were comparable and, in some cases, better than the results of commonly used machine learning algorithms in terms of accuracy, without losing the interpretability.
The structure of social networks strongly affects how different phenomena spread in human society, from the transmission of information to the propagation of contagious diseases. It is well-known that heterogeneous connectivity strongly favors spread, but a precise characterization of the redundancy present in social networks and its effect on the robustness of transmission is still lacking. This gap is addressed by the metric backbone, a weight- and connectivity-preserving subgraph that is sufficient to compute all shortest paths of weighted graphs. This subgraph is obtained via algebraically-principled axioms and does not require statistical sampling based on null-models. We show that the metric backbones of nine contact networks obtained from proximity sensors in a variety of social contexts are generally very small, 49% of the original graph for one and ranging from about 6% to 20% for the others. This reflects a surprising amount of redundancy and reveals that shortest paths on these networks are very robust to random attacks and failures. We also show that the metric backbone preserves the full distribution of shortest paths of the original contact networks—which must include the shortest inter- and intra-community distances that define any community structure—and is a primary subgraph for epidemic transmission based on pure diffusion processes. This suggests that the organization of social contact networks is based on large amounts of shortest-path redundancy which shapes epidemic spread in human populations. Thus, the metric backbone is an important subgraph with regard to epidemic spread, the robustness of social networks, and any communication dynamics that depend on complex network shortest paths.
Author summary
It is through social networks that contagious diseases spread in human populations, as best illustrated by the current pandemic and efforts to contain it. Measuring such networks from human contact data typically results in noisy and dense graphs that need to be simplified for effective analysis, without removal of their essential features. Thus, the identification of a primary subgraph that maintains the social interaction structure and likely transmission pathways is of relevance for studying epidemic spreading phenomena as well as devising intervention strategies to hinder spread. Here we propose and study the metric backbone as an optimal subgraph for sparsification of social contact networks in the study of simple spreading dynamics. We demonstrate that it is a unique, algebraically-principled network subgraph that preserves all shortest paths. We also discover that nine contact networks obtained from proximity sensors in a variety of social contexts contain large amounts of redundant interactions that can be removed with very little impact on community structure and epidemic spread. This reveals that epidemic spread on social networks is very robust to random interaction removal. However, extraction of the metric backbone subgraph reveals which interventions—strategic removal of specific social interactions—are likely to result in maximum impediment to epidemic spread.
In this contribution we define a new integral for interval-valued functions. This integral is an extension of a decomposition integral for real-valued functions. We present two different extensions of this integral. The first one is based on the integral of Aumann that is an extension of Riemann integral for set-valued functions. The second one is based on endpoint-wise interval operations. Interestingly, both of these at a first sight different approaches lead to the same integral. An illustrative example is added, too.
Fuzzy Cognitive Maps are widely applied to support decision making tasks. It is often hard for experts to create the model of a system that provides the required accuracy but simple enough to easily use in practice. In general, it is better to create complex models first, because they can be computationally reduced later until they preserve the required accuracy but become simple enough. Two novel Fuzzy Cognitive Map reduction methods based on K-Means and Fuzzy C-Means clustering are suggested in order to generate simplified models that hopefully mimic the behavior of the original model better than the already existing methods. After the quick overview of the existing techniques found in literature, a simple and a complex model of a real-life problem are reduced to varying degrees with the suggested new methods and with an existing one. The first results of the comparison are published in this paper, too.
Neuroimaging techniques—for example, electroencephalographic (EEG), optical (fNIRS, near infrared spectroscopy), and magnetic resonance imaging (MRI)—make it possible to study brain function and structure, and have made an essential contribution to the diagnosis and treatment of neurological and cognitive problems. The EEG detects electrical activity in the brain using electrodes attached to the scalp. From the electrical signals of the EEG, it is possible to construct maps of the oscillatory brain activity, according to its frequency, for the assessment of brain function during information processing or at rest. fNIRS is an optical neuroimaging technique that measures brain's metabolic response. From the fNIRS signal, it is possible to build brain function maps by measuring the cerebral cortex's optical changes. Magnetic resonance imaging is another neuroimaging technique that uses a magnetic field and radio waves to generate images of the brain's function and structure without using radioactive tracers. In this chapter, these three techniques are discussed. In the first section of this chapter, the electrical signals produced by the brain are reviewed. These signals are involved in many cognitive and behavioral processes, but will only be reviewed when subjects are at rest. The second section of this chapter introduces the fNIRS technique, including the physical and physiological principles underlying optical imaging and the technical considerations for conducting an fNIRS experiment. The last section corresponds to the generation of conventional and functional magnetic resonance imaging (MRI), starting from the basic physical principles and ending with its immediate clinical application.
The probability mixture structure of additive fuzzy systems allows uniform convergence of the generalized probability mixtures that represent the if-then rules of one system or of many combined systems. A new theorem extends this result and shows that it still holds uniformly for any continuous function of such fuzzy systems if the underlying functions are bounded. This allows fuzzy rule-based systems to approximate a far wider range of nonlinear behaviors for a given set of sample data and still produce an explainable probability mixture that governs the rule-based proxy system.
The ever increasing trends of the upcoming technology also lead to many concerns related to environment, society and economics in every field of engineering. Sustainability in any organization implies that something can be sustained indefinitely. Recently, governments and world leaders are advocating for sustainable manufacturing initiatives. To support and promote such initiatives, researchers, scientists, engineers and academic institutions have a responsibility to introduce educational programs related to sustainable manufacturing to prepare the future generation. Advances in sustainable machining are a subset of sustainable manufacturing. Sustainable machining is defined as the creation of products by cutting material that uses the processes which are environmental friendly, economically sound and safe for employees, consumers and communities as well as can conserve energy.
Sustainability in the machining process can be achieved by promoting environmental friendly practices such as dry machining, gas-cooled machining, near dry machining using minimum quantity lubrication, cryogenic cooling, ionic fluids, biodegradable and eco-friendly cutting fluids, etc. Hybridization of two or more such processes is very much capable to achieve the goal of sustainability. Moreover, newly evolving technologies such as use of big data, genetic algorithm, artificial neural network, machine learning and artificial intelligence are the need for machining in the era of industry 4.0. The purpose of this book is to present a group of examples exemplifying research in advances in sustainable machining.
This text provides an in-depth overview of sustainability in machining processes, challenges during machining of difficult-to-cut materials and different ways of green machining in achieving sustainability.
It discusses important topics including green and sustainable machining, dry machining, textured cutting coated tools for machining, solid lubricants-based machining, gas-cooled machining, cryogenic cooling for intelligent machining, artificial neural network for machining, big data based machining, and hybrid intelligent machining.
This book-
Covers advances in sustainable machining such as gas-cooled machining, near dry machining, and minimum quantity lubrication.
Explores use of big data, machine learning and artificial intelligence for machining processes.
Provides case studies and experimental design as well as results with analysis focusing on achieving sustainability.
Discusses artificial intelligence and machine learning based machining processes.
Cover the latest applications of sustainable manufacturing for a better understanding of the concepts.
The text is primarily written for senior undergraduate, graduate students, and researchers in the fields of mechanical, manufacturing, industrial, production engineering and materials science.
The chapter addresses machine intelligence, which refers to the intelligence exhibited by computers. It results from the development of artificial intelligence and is often presented as artificial intelligence. Machine intelligence is basically a prediction technology. Since prediction can have a profound effect on our life, we often make efforts to predict our future from the past and the present. The goal of machine intelligence is giving machines the ability to learn from the changing environment and making predictions.
Penentuan pola musim (curah hujan) di Kabupaten Banyuwangi yaitu dalam penentuan periode-periode musim basah dan musim kering sampai saat ini sangat sulit diprediksi. Hal ini dikarenakan selain perubahan iklimnya sangat drastis dan juga kondisi hidrologis dan faktor topografi di Kabupaten Banyuwangi yang menyebabkan cuaca tidak menentu dan berbeda di daerah satu dengan daerah lainnnya. Begitu pula sebaran stasiun hujan yang tidak merata sehingga menyebabkan data curah hujan tersebut kurang bisa mewakili daerah yang luas. Maksud utama dalam penelitian ini adalah untuk memberikan deskripsi tentang potensi penerapan teori fuzzy di dalam bidang teknik pengembangan sumber daya air yang secara “inherent” dalam penentuan pola musim dari curah hujan di
Kabupaten Banyuwangi untuk pengambilan keputusan pada periode musim basah
ataukah musim kering dalam bentuk aplikasi program Fuzzy Logic.
Secara garis besar langkah-langkah dalam pengerjaan studi ini antara lain adalah pengumpulan data-data sekunder seperti data curah hujan harian, data debit banjir, data klimatologi harian Kabupaten Banyuwangi, melakukan analisa hidrologi, analisa statistik, input data ke aplikasi fuzzy logic, dan validasi output model terhadap data lapangan.
Musim di Kabupaten Banyuwangi berdasarkan hasil analisa curah hujan dengan aplikasi fuzzy logic didapatkan pola musim basah rata-rata terjadi pada bulan Desember–Maret, sedangkan pola musim kering rata-rata terjadi pada bulan Mei-Oktober. Diindikasikan pada bulan April dan Nopember terjadi musim peralihan/pancaroba karena pada beberapa bulan dalam satu tahun yang seharusnya musim kemarau tetapi pada kenyataannya terjadi musim hujan yang diakibatkan oleh penguapan air laut secara besar-besaran.
In this Paper the Imperative reasons for the Problems faced by the IT Professionals using combined overlap block fuzzy cognitive maps (COBFCMS) defined by W.B Vasantha kandasamy is analyzed. Here we are defining the combined overlap block FCM's and it is effective only when the number of concepts can be grouped and are large in numbers. In this paper we analyzed the problems of IT Professionals are large in number. In this the problems of IT Professionals that affect their health are observed and we are finding the reasons for the occurrences of problems. This paper consist of six sections, first section gives the information about development of fuzzy cognitive maps, second section gives preliminaries of fuzzy cognitive maps, and combined overlap block fuzzy cognitive maps, in the third section, we explain about the problems of IT Professionals, in section four we explain the method of determining their hidden pattern, in section five, we give the concept of the problem, final section gives the conclusion based on our studies.
This Paper deals with the systematic study on problems faced by IT Professionals in Chennai using Induced Fuzzy Cognitive Maps (IFCMS).The imperative reasons for the study is to derive an optimistic solution to theproblemwith an unsupervised data collected from a survey among IT Professionals. The problems of IT Professionals are analyzed by a tool called fuzzy theory in general and in particular by induced fuzzy cognitive maps with the help of directed graphs andconnection matrices.
ResearchGate has not been able to resolve any references for this publication.