Showing papers on "Neuro-fuzzy published in 2004"

Fuzzy expert systems and fuzzy reasoning

Abstract: Preface. 1 Introduction. 1.1 Characteristics of Expert Systems. 1.2 Neural Nets. 1.3 Symbolic Reasoning. 1.4 Developing a Rule-Based Expert System. 1.5 Fuzzy Rule-Based Systems. 1.6 Problems in Learning How to Construct Fuzzy Expert Systems. 1.7 Tools for Learning How to Construct Fuzzy Expert Systems. 1.8 Auxiliary Reading. 1.9 Summary. 1.10 Questions. 2 Rule-Based Systems: Overview. 2.1 Expert Knowledge: Rules and Data. 2.2 Rule Antecedent and Consequent. 2.3 Data-Driven Systems. 2.4 Run and Command Modes. 2.5 Forward and Backward Chaining. 2.6 Program Modularization and Blackboard Systems. 2.7 Handling Uncertainties in an Expert System. 2.8 Summary. 2.9 Questions. 3 Fuzzy Logic, Fuzzy Sets, and Fuzzy Numbers: I. 3.1 Classical Logic. 3.2 Elementary Fuzzy Logic and Fuzzy Propositions. 3.3 Fuzzy Sets. 3.4 Fuzzy Relations. 3.5 Truth Value of Fuzzy Propositions. 3.6 Fuzzification and Defuzzification. 3.7 Questions. 4 Fuzzy Logic, Fuzzy Sets, and Fuzzy Numbers: II. 4.1 Introduction. 4.2 Algebra of Fuzzy Sets. 4.3 Approximate Reasoning. 4.4 Hedges. 4.5 Fuzzy Arithmetic. 4.6 Comparisons between Fuzzy Numbers. 4.7 Fuzzy Propositions. 4.8 Questions. 5 Combining Uncertainties. 5.1 Generalizing AND and OR Operators. 5.2 Combining Single Truth Values. 5.3 Combining Fuzzy Numbers and Membership Functions. 5.4 Bayesian Methods. 5.5 The Dempster-Shafer Method. 5.6 Summary. 5.7 Questions. 6 Inference in an Expert System I. 6.1 Overview. 6.2 Types of Fuzzy Inference. 6.3 Nature of Inference in a Fuzzy Expert System. 6.4 Modification and Assignment of Truth Values. 6.5 Approximate Reasoning. 6.6 Tests of Procedures to Obtain the Truth Value of a Consequent from the Truth Value of Its Antecedent. 6.7 Summary. 6.8 Questions. 7 Inference in a Fuzzy Expert System II: Modification of Data and Truth Values. 7.1 Modification of Existing Data by Rule Consequent Instructions. 7.2 Modification of Numeric Discrete Fuzzy Sets: Linguistic Variables and Linguistic Terms. 7.3 Selection of Reasoning Type and Grade-of-Membership Initialization. 7.4 Fuzzification and Defuzzification. 7.5 Non-numeric Discrete Fuzzy Sets. 7.6 Discrete Fuzzy Sets: Fuzziness, Ambiguity, and Contradiction. 7.7 Invalidation of Data: Non-monotonic Reasoning. 7.8 Modification of Values of Data. 7.9 Modeling the Entire Rule Space. 7.10 Reducing the Number of Classification Rules Required in the Conventional Intersection Rule Configuration. 7.11 Summary. 7.12 Questions. 8 Resolving Contradictions: Possibility and Necessity. 8.1 Definition of Possibility and Necessity. 8.2 Possibility and Necessity Suitable for MultiStep Rule-Based Fuzzy Reasoning. 8.3 Modification of Truth Values During a Fuzzy Reasoning Process. 8.4 Formulation of Rules for Possibility and Necessity. 8.5 Resolving Contradictions Using Possibility in a Necessity-Based System. 8.6 Summary. 8.7 Questions. 9 Expert System Shells and the Integrated Development Environment (IDE). 9.1 Overview. 9.2 Help Files. 9.3 Program Editing. 9.4 Running the Program. 9.5 Features of General-Purpose Fuzzy Expert Systems. 9.6 Program Debugging. 9.7 Summary. 9.8 Questions. 10 Simple Example Programs. 10.1 Simple FLOPS Programs. 10.2 Numbers.fps. 10.3 Sum.fps. 10.4 Sum.par. 10.5 Comparison of Serial and Parallel FLOPS. 10.6 Membership Functions, Fuzzification and Defuzzification. 10.7 Summary. 10.8 Questions. 11 Running and Debugging Fuzzy Expert Systems I: Parallel Programs. 11.1 Overview. 11.2 Debugging Tools. 11.3 Debugging Short Simple Programs. 11.4 Isolating the Bug: System Modularization. 11.5 The Debug Run. 11.6 Interrupting the Program for Debug Checks. 11.7 Locating Program Defects with Debug Commands. 11.8 Summary. 11.9 Questions. 12 Running and Debugging Expert Systems II: Sequential Rule-Firing. 12.1 Data Acquisition: From a User Versus Automatically Acquired. 12.2 Ways of Solving a Tree-Search Problem. 12.3 Expert Knowledge in Rules auto1.fps. 12.4 Expert Knowledge in a Database: auto2.fps. 12.5 Other Applications of Sequential Rule Firing. 12.5.1 Missionaries and Cannibals. 12.6 Rules that Make Themselves Refireable: Runaway Programs and Recursion. 12.7 Summary. 12.8 Questions. 13 Solving "What?" Problems when the Answer is Expressed in Words. 13.1 General Methods. 13.2 Iris.par: What Species Is It? 13.3 Echocardiogram Pattern Recognition. 13.4 Schizo.par. 13.5 Discussion. 13.6 Questions. 14 Programs that Can Learn from Experience. 14.1 General Methods. 14.2 Pavlov1.par: Learning by Adding Rules. 14.3 Pavlov2.par: Learning by Adding Facts to Long-Term Memory. 14.4 Defining New Data Elements and New: RULEGEN.FPS. 14.5 Most General Way of Creating New Rules and Data Descriptors. 14.6 Discussion. 14.7 Questions. 15 Running On-Line in Real-Time. 15.1 Overview of On-Line Real-Time Work. 15.2 Input/Output On-Line in Real-Time. 15.3 On-Line Real-Time Processing. 15.4 Types of Rules Useful in Real-Time On-Line Work. 15.5 Memory Management. 15.6 Development of On-Line Real-Time Programs. 15.7 Speeding Up a Program. 15.8 Debugging Real-Time Online Programs. 15.9 Discussion. 15.10 Questions. Appendix. Answers. References. Index.

First Course on Fuzzy Theory and Applications

Abstract: Fuzzy Sets.- The Operation of Fuzzy Set.- Fuzzy Relation and Composition.- Fuzzy Graph and Relation.- Fuzzy Number.- Fuzzy Function.- Probabilisy and Uncertainty.- Fuzzy Logic.- Fuzzy Inference.- Fuzzy Control and Fuzzy Expert Systems.- Fusion of Fuzzy System and Neural Networks.- Fusion of Fuzzy Systems and Genetic Algorithms.

Fuzzy Randomness: Uncertainty in Civil Engineering and Computational Mechanics

Abstract: 1 Introduction.- 2 Mathematical Basics for the Formal Description of Uncertainty.- 3 Description of Uncertain Structural Parameters as Fuzzy Variables.- 4 Description of Uncertain Structural Parameters as Fuzzy Random Variables.- 5 Fuzzy and Fuzzy Stochastic Structural Analysis.- 6 Fuzzy Probabilistic Safety Assessment.- 7 Structural Design Based on Clustering.- References.

Modeling complex systems using fuzzy cognitive maps

Abstract: This research deals with the soft computing methodology of fuzzy cognitive map (FCM). Here a mathematical description of FCM is presented and a new methodology based on fuzzy logic techniques for developing the FCM is examined. The capability and usefulness of FCM in modeling complex systems and the application of FCM to modeling and describing the behavior of a heat exchanger system is presented. The applicability of FCM to model the supervisor of complex systems is discussed and the FCM-supervisor for evaluating the performance of a system is constructed; simulation results are presented and discussed.

Foundations of soft case-based reasoning

Abstract: FOREWORD.PREFACE.ABOUT THE AUTHORS.1 INTRODUCTION.1.1 Background.1.2 Components and Features of Case-Based Reasoning.1.2.1 CBR System versus Rule-Based System.1.2.2 CBR versus Human Reasoning.1.2.3 CBR Life Cycle.1.3 Guidelines for the Use of Case-Based Reasoning.1.4 Advantages of Using Case-Based Reasoning.1.5 Case Representation and Indexing.1.5.1 Case Representation.1.5.2 Case Indexing.1.6 Case Retrieval.1.7 Case Adaptation.1.8 Case Learning and Case-Base Maintenance.1.8.1 Learning in CBR Systems.1.8.2 Case-Base Maintenance.1.9 Example of Building a Case-Based Reasoning System.1.9.1 Case Representation.1.9.2 Case Indexing.1.9.3 Case Retrieval.1.9.4 Case Adaptation.1.9.5 Case-Base Maintenance.1.10 Case-Based Reasoning: Methodology or Technology?1.11 Soft Case-Based Reasoning.1.11.1 Fuzzy Logic.1.11.2 Neural Networks.1.11.3 Genetic Algorithms.1.11.4 Some CBR Tasks for Soft Computing Applications.1.12 Summary.References.2 CASE REPRESENTATION AND INDEXING.2.1 Introduction.2.2 Traditional Methods of Case Representation.2.2.1 Relational Representation.2.2.2 Object-Oriented Representation.2.2.3 Predicate Representation.2.2.4 Comparison of Case Representations.2.3 Soft Computing Techniques for Case Representation.2.3.1 Case Knowledge Representation Based on Fuzzy Sets.2.3.2 Rough Sets and Determining Reducts.2.3.3 Prototypical Case Generation Using Reducts with Fuzzy Representation.2.4 Case Indexing.2.4.1 Traditional Indexing Method.2.4.2 Case Indexing Using a Bayesian Model.2.4.3 Case Indexing Using a Prototype-Based Neural Network.2.4.4 Case Indexing Using a Three-Layered Back Propagation Neural Network.2.5 Summary.References.3 CASE SELECTION AND RETRIEVAL.3.1 Introduction.3.2 Similarity Concept.3.2.1 Weighted Euclidean Distance.3.2.2 Hamming and Levenshtein Distances.3.2.3 Cosine Coefficient for Text-Based Cases.3.2.4 Other Similarity Measures.3.2.5 k-Nearest Neighbor Principle.3.3 Concept of Fuzzy Sets in Measuring Similarity.3.3.1 Relevance of Fuzzy Similarity in Case Matching.3.3.2 Computing Fuzzy Similarity Between Cases.3.4 Fuzzy Classification and Clustering of Cases.3.4.1 Weighted Intracluster and Intercluster Similarity.3.4.2 Fuzzy ID3 Algorithm for Classification.3.4.3 Fuzzy c-Means Algorithm for Clustering.3.5 Case Feature Weighting.3.5.1 Using Gradient-Descent Technique and Neural Networks.3.5.2 Using Genetic Algorithms.3.6 Case Selection and Retrieval Using Neural Networks.3.6.1 Methodology.3.6.2 Glass Identification.3.7 Case Selection Using a Neuro-Fuzzy Model.3.7.1 Selection of Cases and Class Representation.3.7.2 Formulation of the Network.3.8 Case Selection Using Rough-Self Organizing Map.3.8.1 Pattern Indiscernibility and Fuzzy Discretization of Feature Space.3.8.2 Methodology for Generation of Reducts.3.8.3 Rough SOM.3.8.4 Experimental Results.3.9 Summary.References.4 CASE ADAPTATION.4.1 Introduction.4.2 Traditional Case Adaptation Strategies.4.2.1 Reinstantiation.4.2.2 Substitution.4.2.3 Transformation.4.2.4 Example of Adaptation Knowledge in Pseudocode.4.3 Some Case Adaptation Methods.4.3.1 Learning Adaptation Cases.4.3.2 Integrating Rule- and Case-Based Adaptation Approaches.4.3.3 Using an Adaptation Matrix.4.3.4 Using Configuration Techniques.4.4 Case Adaptation Through Machine Learning.4.4.1 Fuzzy Decision Tree.4.4.2 Back-Propagation Neural Network.4.4.3 Bayesian Model.4.4.4 Support Vector Machine.4.4.5 Genetic Algorithms.4.5 Summary.References.5 CASE-BASE MAINTENANCE.5.1 Introduction.5.2 Background.5.3 Types of Case-Base Maintenance.5.3.1 Qualitative Maintenance.5.3.2 Quantitative Maintenance.5.4 Case-Base Maintenance Using a Rough-Fuzzy Approach.5.4.1 Maintaining the Client Case Base.5.4.2 Experimental Results.5.4.3 Complexity Issues.5.5 Case-Base Maintenance Using a Fuzzy Integral Approach.5.5.1 Fuzzy Measures and Fuzzy Integrals.5.5.2 Case-Base Competence.5.5.3 Fuzzy Integral-Based Competence Model.5.5.4 Experiment Results.5.6 Summary.References.6 APPLICATIONS.6.1 Introduction.6.2 Web Mining.6.2.1 Case Representation Using Fuzzy Sets.6.2.2 Mining Fuzzy Association Rules.6.3 Medical Diagnosis.6.3.1 System Architecture.6.3.2 Case Retrieval Using a Fuzzy Neural Network.6.3.3 Case Evaluation and Adaptation Using Induction.6.4 Weather Prediction.6.4.1 Structure of the Hybrid CBR System.6.4.2 Case Adaptation Using ANN.6.5 Legal Inference.6.5.1 Fuzzy Logic in Case Representation.6.5.2 Fuzzy Similarity in Case Retrieval and Inference.6.6 Property Valuation.6.6.1 PROFIT System.6.6.2 Fuzzy Preference in Case Retrieval.6.7 Corporate Bond Rating.6.7.1 Structure of a Hybrid CBR System Using Gas.6.7.2 GA in Case Indexing and Retrieval.6.8 Color Matching.6.8.1 Structure of the Color-Matching Process.6.8.2 Fuzzy Case Retrieval.6.9 Shoe Design.6.9.1 Feature Representation.6.9.2 Neural Networks in Retrieval.6.10 Other Applications.6.11 Summary.References.APPENDIXES.A FUZZY LOGIC.A.1 Fuzzy Subsets.A.2 Membership Functions.A.3 Operations on Fuzzy Subsets.A.4 Measure of Fuzziness.A.5 Fuzzy Rules.A.5.1 Definition.A.5.2 Fuzzy Rules for Classification.References.B ARTIFICIAL NEURAL NETWORKS.B.1 Architecture of Artificial Neural Networks.B.2 Training of Artificial Neural Networks.B.3 ANN Models.B.3.1 Single-Layered Perceptron.B.3.2 Multilayered Perceptron Using a Back-Propagation Algorithm.B.3.3 Radial Basis Function Network.B.3.4 Kohonen Neural Network.References.C GENETIC ALGORITHMS.C.1 Basic Principles.C.2 Standard Genetic Algorithm.C.3 Examples.C.3.1 Function Maximization.C.3.2 Traveling Salesman Problem.References.D ROUGH SETS.D.1 Information Systems.D.2 Indiscernibility Relation.D.3 Set Approximations.D.4 Rough Membership.D.5 Dependency of Attributes.References.INDEX.

Flexible neuro-fuzzy systems

Abstract: In this paper, we derive new neuro-fuzzy structures called flexible neuro-fuzzy inference systems or FLEXNFIS. Based on the input-output data, we learn not only the parameters of the membership functions but also the type of the systems (Mamdani or logical). Moreover, we introduce: 1) softness to fuzzy implication operators, to aggregation of rules and to connectives of antecedents; 2) certainty weights to aggregation of rules and to connectives of antecedents; and 3) parameterized families of T-norms and S-norms to fuzzy implication operators, to aggregation of rules and to connectives of antecedents. Our approach introduces more flexibility to the structure and design of neuro-fuzzy systems. Through computer simulations, we show that Mamdani-type systems are more suitable to approximation problems, whereas logical-type systems may be preferred for classification problems.

Fuzzy identification using fuzzy neural networks with stable learning algorithms

Abstract: In general, fuzzy neural networks cannot match nonlinear systems exactly. Unmodeled dynamic leads parameters drift and even instability problem. According to system identification theory, robust modification terms must be included in order to guarantee Lyapunov stability. This paper suggests new learning laws for Mamdani and Takagi-Sugeno-Kang type fuzzy neural networks based on input-to-state stability approach. The new learning schemes employ a time-varying learning rate that is determined from input-output data and model structure. Stable learning algorithms for the premise and the consequence parts of fuzzy rules are proposed. The calculation of the learning rate does not need any prior information such as estimation of the modeling error bounds. This offer an advantage compared to other techniques using robust modification.

Support vector learning mechanism for fuzzy rule-based modeling: a new approach

Abstract: This paper describes a fuzzy modeling framework based on support vector machine, a rule-based framework that explicitly characterizes the representation in fuzzy inference procedure. The support vector learning mechanism provides an architecture to extract support vectors for generating fuzzy IF-THEN rules from the training data set, and a method to describe the fuzzy system in terms of kernel functions. Thus, it has the inherent advantage that the model does not have to determine the number of rules in advance, and the overall fuzzy inference system can be represented as series expansion of fuzzy basis functions. The performance of the proposed approach is compared to other fuzzy rule-based modeling methods using four data sets.

Neuro Fuzzy Systems: Sate-of-the-Art Modeling Techniques

Abstract: Fusion of Artificial Neural Networks (ANN) and Fuzzy Inference Systems (FIS) have attracted the growing interest of researchers in various scientific and engineering areas due to the growing need of adaptive intelligent systems to solve the real world problems. ANN learns from scratch by adjusting the interconnections between layers. FIS is a popular computing framework based on the concept of fuzzy set theory, fuzzy if-then rules, and fuzzy reasoning. The advantages of a combination of ANN and FIS are obvious. There are several approaches to integrate ANN and FIS and very often it depends on the application. We broadly classify the integration of ANN and FIS into three categories namely concurrent model, cooperative model and fully fused model. This paper starts with a discussion of the features of each model and generalize the advantages and deficiencies of each model. We further focus the review on the different types of fused neuro-fuzzy systems and citing the advantages and disadvantages of each model.

A neuro-fuzzy scheme for simultaneous feature selection and fuzzy rule-based classification

Abstract: Most methods of classification either ignore feature analysis or do it in a separate phase, offline prior to the main classification task. This paper proposes a neuro-fuzzy scheme for designing a classifier along with feature selection. It is a four-layered feed-forward network for realizing a fuzzy rule-based classifier. The network is trained by error backpropagation in three phases. In the first phase, the network learns the important features and the classification rules. In the subsequent phases, the network is pruned to an "optimal" architecture that represents an "optimal" set of rules. Pruning is found to drastically reduce the size of the network without degrading the performance. The pruned network is further tuned to improve performance. The rules learned by the network can be easily read from the network. The system is tested on both synthetic and real data sets and found to perform quite well.

Combining rough and fuzzy sets for feature selection

Abstract: Feature selection (FS) refers to the problem of selecting those input attributes that are most predictive of a given outcome; a problem encountered in many areas such as machine learning, pattern recognition and signal processing. Unlike other dimensionality reduction methods, feature selectors preserve the original meaning of the features after reduction. This has found application in tasks that involve datasets containing huge numbers of features (in the order of tens of thousands), which would be impossible to process further. Recent examples include text processing and web content classification. FS techniques have also been applied to small and medium-sized datasets in order to locate the most informative features for later use. Many feature selection methods have been developed and are reviewed critically in this thesis, with particular emphasis on their current limitations. The leading methods in this field are presented in a consistent algorithmic framework. One of the many successful applications of rough set theory has been to this area. The rough set ideology of using only the supplied data and no other information has many benefits in FS, where most other methods require supplementary knowledge. However, the main limitation of rough set-based feature selection in the literature is the restrictive requirement that all data is discrete. In classical rough set theory, it is not possible to consider real-valued or noisy data. This thesis proposes and develops an approach based on fuzzy-rough sets, fuzzy rough feature selection (FRFS), that addresses these problems and retains dataset semantics. Complexity analysis of the underlying algorithms is included. FRFS is applied to two domains where a feature reducing step is important; namely, web content classification and complex systems monitoring. The utility of this approach is demonstrated and is compared empirically with several dimensionality reducers. In the experimental studies, FRFS is shown to equal or improve classification accuracy when compared to the results from unreduced data. Classifiers that use a lower dimensional set of attributes which are retained by fuzzy-rough reduction outperform those that employ more attributes returned by the existing crisp rough reduction method. In addition, it is shown that FRFS is more powerful than the other FS techniques in the comparative study.

A Hybrid Neuro-Fuzzy Model for Mineral Potential Mapping

Abstract: A GIS-based hybrid neuro-fuzzy approach to mineral potential mapping implements a Takagi–Sugeno type fuzzy inference system in a four-layered feed-forward adaptive neural network. In this approach, each unique combination of predictor patterns is considered a feature vector whose components are derived by knowledge-based ordinal encoding of the constituent predictor patterns. A subset of feature vectors with a known output target vector (i.e., unique conditions known to be associated with either a mineralized or a barren location), extracted from a set of all feature vectors, is used for the training of an adaptive neuro-fuzzy inference system. Training involves iterative adjustment of parameters of the adaptive neuro-fuzzy inference system using a hybrid learning procedure for mapping each training vector to its output target vector with minimum sum of squared error. The trained adaptive neuro-fuzzy inference system is used to process all feature vectors. The output for each feature vector is a value that indicates the extent to which a feature vector belongs to the mineralized class or the barren class. These values are used to generate a favorability map. The procedure is applied to regional-scale base metal potential mapping in a study area located in the Aravalli metallogenic province (western India). The adaptive neuro-fuzzy inference system demarcates high favorability zones occupying 9.75% of the study area and identifies 96% of the known base metal deposits. This result is significant both in terms of reduction in search area and the percentage of deposits identified.

Prediction and identification using wavelet-based recurrent fuzzy neural networks

Abstract: This paper presents a wavelet-based recurrent fuzzy neural network (WRFNN) for prediction and identification of nonlinear dynamic systems. The proposed WRFNN model combines the traditional Takagi-Sugeno-Kang (TSK) fuzzy model and the wavelet neural networks (WNN). This paper adopts the nonorthogonal and compactly supported functions as wavelet neural network bases. Temporal relations embedded in the network are caused by adding some feedback connections representing the memory units into the second layer of the feedforward wavelet-based fuzzy neural networks (WFNN). An online learning algorithm, which consists of structure learning and parameter learning, is also presented. The structure learning depends on the degree measure to obtain the number of fuzzy rules and wavelet functions. Meanwhile, the parameter learning is based on the gradient descent method for adjusting the shape of the membership function and the connection weights of WNN. Finally, computer simulations have demonstrated that the proposed WRFNN model requires fewer adjustable parameters and obtains a smaller RMS error than other methods.

Online tuning of fuzzy inference systems using dynamic fuzzy Q-learning

Abstract: This paper presents a dynamic fuzzy Q-learning (DFQL) method that is capable of tuning fuzzy inference systems (FIS) online. A novel online self-organizing learning algorithm is developed so that structure and parameters identification are accomplished automatically and simultaneously based only on Q-learning. Self-organizing fuzzy inference is introduced to calculate actions and Q-functions so as to enable us to deal with continuous-valued states and actions. Fuzzy rules provide a natural mean of incorporating the bias components for rapid reinforcement learning. Experimental results and comparative studies with the fuzzy Q-learning (FQL) and continuous-action Q-learning in the wall-following task of mobile robots demonstrate that the proposed DFQL method is superior.

Determining the importance weights for the design requirements in the house of quality using the fuzzy analytic network approach: The Fuzzy Analytic Network Approach

Abstract: Quality function deployment (QFD) has been used to translate customer needs (CNs) and wants into technical design requirements (DRs) in order to increase customer satisfaction. QFD uses the house of quality (HOQ), which is a matrix providing a conceptual map for the design process, as a construct for understanding CNs and establishing priorities of DRs to satisfy them. This article uses the analytic network process (ANP), the general form of the analytic hierarchy process (AHP), to prioritize DRs by taking into account the degree of the interdependence between the CNs and DRs and the inner dependence among them. In addition, because human judgment on the importance of requirements is always imprecise and vague, this work concentrates on a fuzzy ANP approach in which triangular fuzzy numbers are used to improve the quality of the responsiveness to CNs and DRs. A numerical example is presented to show the proposed methodology. © 2004 Wiley Periodicals, Inc.

Adaptive neuro-fuzzy intrusion detection systems

Abstract: The Intrusion Detection System architecture commonly used in commercial and research systems have a number of problems that limit their configurability, scalability or efficiency. In this paper, two machine-learning paradigms, Artificial Neural Networks and Fuzzy Inference System, are used to design an Intrusion Detection System. SNORT is used to perform real time traffic analysis and packet logging on IP network during the training phase of the system. Then a signature pattern database is constructed using protocol analysis and Neuro-Fuzzy learning method. Using 1998 DARPA Intrusion Detection Evaluation Data and TCP dump raw data, the experiments are deployed and discussed.

Generating an interpretable family of fuzzy partitions from data

Abstract: In this paper, we propose a new method to construct fuzzy partitions from data. The procedure generates a hierarchy including best partitions of all sizes from n to two fuzzy sets. The maximum size n is determined according to the data distribution and corresponds to the finest resolution level. We use an ascending method for which a merging criterion is needed. This criterion is based on the definition of a special metric distance suitable for fuzzy partitioning, and the merging is done under semantic constraints. The distance we define does not handle the point coordinates, but directly their membership degrees to the fuzzy sets of the partition. This leads to the introduction of the notions of internal and external distances. The hierarchical fuzzy partitioning is carried independently over each dimension, and, to demonstrate the partition potential, they are used to build fuzzy inference system using a simple selection mechanism. Due to the merging technique, all the fuzzy sets in the various partitions are interpretable as linguistic labels. The tradeoff between accuracy and interpretability constitutes the most promising aspect in our approach. Well known data sets are investigated and the results are compared with those obtained by other authors using different techniques. The method is also applied to real world agricultural data, the results are analyzed and weighed against those achieved by other methods, such as fuzzy clustering or discriminant analysis.

Fuzzy Neural Network Theory and Application

Abstract: Fuzzy Neural Networks for Storing and Classifying Feedback Fuzzy Associative Memory Regular Fuzzy Neural Networks Polygonal Fuzzy Neural Networks Approximation Analysis of Fuzzy Systems Stochastic Fuzzy Systems and Approximation Application of Fuzzy Neural Networks to Image Restoration

A Neuro-Fuzzy Approach for Modelling Electricity Demand in Victoria

Abstract: Neuro-fuzzy systems have attracted growing interest of researchers in various scientific and engineering areas due to the increasing need of intelligent systems. This paper evaluates the use of two popular soft computing techniques and conventional statistical approach based on Box--Jenkins autoregressive integrated moving average (ARIMA) model to predict electricity demand in the State of Victoria, Australia. The soft computing methods considered are an evolving fuzzy neural network (EFuNN) and an artificial neural network (ANN) trained using scaled conjugate gradient algorithm (CGA) and backpropagation (BP) algorithm. The forecast accuracy is compared with the forecasts used by Victorian Power Exchange (VPX) and the actual energy demand. To evaluate, we considered load demand patterns for 10 consecutive months taken every 30 min for training the different prediction models. Test results show that the neuro-fuzzy system performed better than neural networks, ARIMA model and the VPX forecasts.

Adaptive neuro-fuzzy intrusion detection systems

Abstract: The intrusion detection system architecture commonly used in commercial and research systems have a number of problems that limit their configurability, scalability or efficiency. In this paper, two machine-learning paradigms, artificial neural networks and fuzzy inference system, are used to design an intrusion detection system. SNORT is used to perform real time traffic analysis and packet logging on IP network during the training phase of the system. Then a signature pattern database is constructed using protocol analysis and neuro-fuzzy learning method. Using 1998 DARPA Intrusion Detection Evaluation Data and TCP dump raw data, the experiments are deployed and discussed.