基于遗传算法的多维模糊分类器构造的研究

讨论了基于模糊遗传机器学习机制的密歇根方法在多维分类问题上的应用及性能问题,并提出了一种新的模糊遗传学习方法.将每一模糊规则作为遗传算法中的一个个体,且具有相应的适应度函数值.在提取模糊规则的同时,还对每个属性维的模糊划分进行学习以获取较好的模糊集合参数.另外,该方法引入了基于相似性的选择机制,减轻了选择机制对低适应函数值个体造成的选择压力,保持了种群的多样性,从而有效地避免了遗传算法收敛到局部解的问题.实验结果表明,该方法在多维模糊分类器的构造问题上具有较高的正确分类率、适应性较好等性能.     

Performance evaluation of fuzzy rule-based classification systems obtained by multi-objective genetic algorithms

In this paper, we examine the classification performance of fuzzy if-then rules selected by a GA-based multi-objective rule selection method. This rule selection method can be applied to high-dimensional pattern classification problems with many continuous attributes by restricting the number of antecedent conditions of each candidate fuzzy if-then rule. As candidate rules, we only use fuzzy if-then rules with a small number of antecedent conditions. Thus it is easy for human users to understand each rule selected by our method. Our rule selection method has two objectives: to minimize the number of selected fuzzy if-then rules and to maximize the number of correctly classified patterns. In our multi-objective fuzzy rule selection problem, there exist several solutions (i.e., several rule sets) called “non-dominated solutions” because two conflicting objectives are considered. In this paper, we examine the performance of our GA-based rule selection method by computer simulations on a real-world pattern classification problem with many continuous attributes. First we examine the classification performance of our method for training patterns by computer simulations. Next we examine the generalization ability for test patterns. We show that a fuzzy rule-based classification system with an appropriate number of rules has high generalization ability.     

An approach to fuzzy default reasoning for function approximation

This paper discusses fuzzy reasoning for approximately realizing nonlinear functions by a small number of fuzzy if-then rules with different specificity levels. Our fuzzy rule base is a mixture of general and specific rules, which overlap with each other in the input space. General rules work as default rules in our fuzzy rule base. First, we briefly describe existing approaches to the handling of default rules in the framework of possibility theory. Next, we show that standard interpolation-based fuzzy reasoning leads to counterintuitive results when general rules include specific rules with different consequents. Then, we demonstrate that intuitively acceptable results are obtained from a non-standard inclusion-based fuzzy reasoning method. Our approach is based on the preference for more specific rules, which is a commonly used idea in the field of default reasoning. When a general rule includes a specific rule and they are both compatible with an input vector, the weight of the general rule is discounted in fuzzy reasoning. We also discuss the case where general rules do not perfectly but partially include specific rules. Then we propose a genetics-based machine learning (GBML) algorithm for extracting a small number of fuzzy if-then rules with different specificity levels from numerical data using our inclusion-based fuzzy reasoning method. Finally, we describe how our approach can be applied to the approximate realization of fuzzy number-valued nonlinear functions     

Effect of rule weights in fuzzy rule-based classification systems

This paper examines the effect of rule weights in fuzzy rule-based classification systems. Each fuzzy IF-THEN rule in our classification system has antecedent linguistic values and a single consequent class. We use a fuzzy reasoning method based on a single winner rule in the classification phase. The winner rule for a new pattern is the fuzzy IF-THEN rule that has the maximum compatibility grade with the new pattern. When we use fuzzy IF-THEN rules with certainty grades, the winner is determined as the rule with the maximum product of the compatibility grade and the certainty grade. In this paper, the effect of rule weights is illustrated by drawing classification boundaries using fuzzy IF-THEN rules with/without certainty grades. It is also shown that certainty grades play an important role when a fuzzy rule-based classification system is a mixture of general rules and specific rules. Through computer simulations, we show that comprehensible fuzzy rule-based systems with high classification performance can be designed without modifying the membership functions of antecedent linguistic values when we use fuzzy IF-THEN rules with certainty grades     

An adaptive neural fuzzy filter and its applications

A new kind of nonlinear adaptive filter, the adaptive neural fuzzy filter (ANFF), based upon a neural network's learning ability and fuzzy if-then rule structure, is proposed in this paper. The ANFF is inherently a feedforward multilayered connectionist network which can learn by itself according to numerical training data or expert knowledge represented by fuzzy if-then rules. The adaptation here includes the construction of fuzzy if-then rules (structure learning), and the tuning of the free parameters of membership functions (parameter learning). In the structure learning phase, fuzzy rules are found based on the matching of input-output clusters. In the parameter learning phase, a backpropagation-like adaptation algorithm is developed to minimize the output error. There are no hidden nodes (i.e., no membership functions and fuzzy rules) initially, and both the structure learning and parameter learning are performed concurrently as the adaptation proceeds. However, if some linguistic information about the design of the filter is available, such knowledge can be put into the ANFF to form an initial structure with hidden nodes. Two major advantages of the ANFF can thus be seen: 1) a priori knowledge can be incorporated into the ANFF which makes the fusion of numerical data and linguistic information in the filter possible; and 2) no predetermination, like the number of hidden nodes, must be given, since the ANFF can find its optimal structure and parameters automatically     

A neural fuzzy system with fuzzy supervised learning

A neural fuzzy system learning with fuzzy training data (fuzzy if-then rules) is proposed in this paper. This system is able to process and learn numerical information as well as linguistic information. At first, we propose a five-layered neural network for the connectionist realization of a fuzzy inference system. The connectionist structure can house fuzzy logic rules and membership functions for fuzzy inference. We use alpha-level sets of fuzzy numbers to represent linguistic information. The inputs, outputs, and weights of the proposed network can be fuzzy numbers of any shape. Furthermore, they can be hybrid of fuzzy numbers and numerical numbers through the use of fuzzy singletons. Based on interval arithmetics, a fuzzy supervised learning algorithm is developed for the proposed system. It extends the normal supervised learning techniques to the learning problems where only linguistic teaching signals are available. The fuzzy supervised learning scheme can train the proposed system with desired fuzzy input-output pairs which are fuzzy numbers instead of the normal numerical values. With fuzzy supervised learning, the proposed system can be used for rule base concentration to reduce the number of rules in a fuzzy rule base. Simulation results are presented to illustrate the performance and applicability of the proposed system.     

An ensemble method for fuzzy rule-based classification systems

Fuzzy rule-based classification systems are very useful tools in the field of machine learning as they are able to build linguistic comprehensible models. However, these systems suffer from exponential rule explosion when the number of variables increases, degrading, therefore, the accuracy of these systems as well as their interpretability. In this article, we propose to improve the comprehensibility through a supervised learning method by automatic generation of fuzzy classification rules, designated SIFCO–PAF. Our method reduces the complexity by decreasing the number of rules and of antecedent conditions, making it thus adapted to the representation and the prediction of rather high-dimensional pattern classification problems. We perform, firstly, an ensemble methodology by combining a set of simple classification models. Subsequently, each model uses a subset of the initial attributes: In this case, we propose to regroup the attributes using linear correlation search among the training set elements. Secondly, we implement an optimal fuzzy partition thanks to supervised discretization followed by an automatic membership functions construction. The SIFCO–PAF method, analyzed experimentally on various data sets, guarantees an important reduction in the number of rules and of antecedents without deteriorating the classification rates, on the contrary accuracy is even improved.     

Hybridization of fuzzy GBML approaches for pattern classification problems

We propose a hybrid algorithm of two fuzzy genetics-based machine learning approaches (i.e., Michigan and Pittsburgh) for designing fuzzy rule-based classification systems. First, we examine the search ability of each approach to efficiently find fuzzy rule-based systems with high classification accuracy. It is clearly demonstrated that each approach has its own advantages and disadvantages. Next, we combine these two approaches into a single hybrid algorithm. Our hybrid algorithm is based on the Pittsburgh approach where a set of fuzzy rules is handled as an individual. Genetic operations for generating new fuzzy rules in the Michigan approach are utilized as a kind of heuristic mutation for partially modifying each rule set. Then, we compare our hybrid algorithm with the Michigan and Pittsburgh approaches. Experimental results show that our hybrid algorithm has higher search ability. The necessity of a heuristic specification method of antecedent fuzzy sets is also demonstrated by computational experiments on high-dimensional problems. Finally, we examine the generalization ability of fuzzy rule-based classification systems designed by our hybrid algorithm.     

Neural networks that learn from fuzzy if-then rules

An architecture for neural networks that can handle fuzzy input vectors is proposed, and learning algorithms that utilize fuzzy if-then rules as well as numerical data in neural network learning for classification problems and for fuzzy control problems are derived. The learning algorithms can be viewed as an extension of the backpropagation algorithm to the case of fuzzy input vectors and fuzzy target outputs. Using the proposed methods, linguistic knowledge from human experts represented by fuzzy if-then rules and numerical data from measuring instruments can be integrated into a single information processing system (classification system or fuzzy control system). It is shown that the scheme works well for simple examples     

基于数值型数据的模糊规则快速挖掘方法

针对数值型数据能准确反应现实世界,但难以理解的问题,为了从数值型数据中挖掘出易于理解的知识,提出了基于数值型数据的模糊规则快速挖掘方法。该方法能从数值型数据中挖掘出一个零阶的Sugeno模糊规则,并采用一种启发式方法将这个零阶的Sugeno模糊规则的数值结论转变为两个带置信度的语言变量,并给出了规则库的存储结构。最后通过实例证明了这种快速模糊规则挖掘方法能避免复杂的数值型计算和能有效逼近非线性函数的优点．     

On the representation, measurement, and discovery of fuzzy associations

The use of fuzzy sets to describe associations between data extends the types of relationships that may be represented, facilitates the interpretation of rules in linguistic terms, and avoids unnatural boundaries in the partitioning of the attribute domains. In addition, the partial membership values provide a method for incorporating the distribution of the data into the assessment of a rule. This paper investigates techniques to identify and evaluate associations in a relational database that are expressible by fuzzy if-then rules. Extensions of the classical confidence measure based on the /spl alpha/-cut decompositions of the fuzzy sets are proposed to incorporate the distribution of the data into the assessment of a relationship and identify robustness in an association. A rule learning strategy that discovers both the presence and the type of an association is presented.     

Selecting fuzzy if-then rules for classification problems usinggenetic algorithms

This paper proposes a genetic-algorithm-based method for selecting a small number of significant fuzzy if-then rules to construct a compact fuzzy classification system with high classification power. The rule selection problem is formulated as a combinatorial optimization problem with two objectives: to maximize the number of correctly classified patterns and to minimize the number of fuzzy if-then rules. Genetic algorithms are applied to this problem. A set of fuzzy if-then rules is coded into a string and treated as an individual in genetic algorithms. The fitness of each individual is specified by the two objectives in the combinatorial optimization problem. The performance of the proposed method for training data and test data is examined by computer simulations on the iris data of Fisher     

Support-vector-based fuzzy neural network for pattern classification

Fuzzy neural networks (FNNs) for pattern classification usually use the backpropagation or C-cluster type learning algorithms to learn the parameters of the fuzzy rules and membership functions from the training data. However, such kinds of learning algorithms usually cannot minimize the empirical risk (training error) and expected risk (testing error) simultaneously, and thus cannot reach a good classification performance in the testing phase. To tackle this drawback, a support-vector-based fuzzy neural network (SVFNN) is proposed for pattern classification in this paper. The SVFNN combines the superior classification power of support vector machine (SVM) in high dimensional data spaces and the efficient human-like reasoning of FNN in handling uncertainty information. A learning algorithm consisting of three learning phases is developed to construct the SVFNN and train its parameters. In the first phase, the fuzzy rules and membership functions are automatically determined by the clustering principle. In the second phase, the parameters of FNN are calculated by the SVM with the proposed adaptive fuzzy kernel function. In the third phase, the relevant fuzzy rules are selected by the proposed reducing fuzzy rule method. To investigate the effectiveness of the proposed SVFNN classification, it is applied to the Iris, Vehicle, Dna, Satimage, Ijcnn1 datasets from the UCI Repository, Statlog collection and IJCNN challenge 2001, respectively. Experimental results show that the proposed SVFNN for pattern classification can achieve good classification performance with drastically reduced number of fuzzy kernel functions.     

Learning Premises of Fuzzy Rules for Knowledge Acquisition in Classification Problems

A key issue in building fuzzy classification systems is the specification of rule conditions, which determine the structure of a knowledge base. This paper presents a new approach to automatically extract classification knowledge from numerical data by means of premise learning. A genetic algorithm is employed to search for premise structure in combination with parameters of membership functions of input fuzzy sets to yield optimal conditions of classification rules. The major advantage of our work is that a parsimonious knowledge base with a low number of rules can be achieved. The practical applicability of the proposed method is examined by computer simulations on two well-known benchmark problems of Iris Data and Cancer Data classification. Received 11 February 1999 / Revised 13 January 2001 / Accepted in revised form 13 February 2001     

AUTOMATICALLY CONSTRUCTING MEMBERSHIP FUNCTIONS AND GENERATING FUZZY RULES USING GENETIC ALGORITHMS

The most important task in designing a fuzzy classification system is to find a set of fuzzy rules from training data to deal with a specific classification problem. In recent years, many methods have been proposed to construct membership functions and generate fuzzy rules from training data for handling fuzzy classification problems. We propose a new method to generate fuzzy rules from training data by using genetic algorithms (GAs). First, we divide the training data into several clusters by using the weighted distance clustering method and generate a fuzzy rule for each cluster. Then, we use GAs to tune the membership functions of the generated fuzzy rules. The proposed method attains a higher average classification accuracy rate than the existing methods.     

可理解模糊分类系统的分层演化学习

针对模糊系统的可理解性要求,结合微粒群算法和遗传算法各自的演化特点,采用两阶段学习策略,对模糊分类系统进行分层演化。首先利用微粒群算法优化各输入变量的语言值数目及对应的模糊集参数,形成候选规则集,再应用遗传算法选择规则,得到可理解的和精确的模糊分类系统。该方法几乎无需先验知识,可直接从实值数据获取模糊分类系统,应用典型分类问题为例说明其有效性。     

Parallelizing fuzzy rule generation using GPGPU

This article proposes a method to parallelize the process of generating fuzzy if-then rules for pattern classification problems in order to reduce the computational time. The proposed method makes use of general purpose computation on graphics processing units (GPGPUs)’ parallel implementation with compute unified device architecture (CUDA), a development environment. CUDA contains a library to perform matrix operations in parallel. In the proposed method, published source codes of matrix multiplication are modified so that the membership values of given training patterns with antecedent fuzzy sets are calculated. In a series of computational experiments, it is shown that the computational time is reduced for those problems that require high computational effort.     

Influential rule search scheme (IRSS) - a new fuzzy pattern classifier

Automatic generation of fuzzy rule base and membership functions from an input-output data set, for reliable construction of an adaptive fuzzy inference system, has become an important area of research interest. We propose a new robust, fast acting adaptive fuzzy pattern classification scheme, named influential rule search scheme (IRSS). In IRSS, rules which are most influential in contributing to the error produced by the adaptive fuzzy system are identified at the end of each epoch and subsequently modified for satisfactory performance. This fuzzy rule base adjustment scheme is accompanied by an output membership function adaptation scheme for fine tuning the fuzzy system architecture. This iterative method has shown a relatively high speed of convergence. Performance of the proposed IRSS is compared with other existing pattern classification schemes by implementing it for Fisher's iris data problem and Wisconsin breast cancer data problems.     

Evolutionary learning of rule premises for fuzzy modelling

The task of fuzzy modelling involves specification of rule antecedents and determination of their consequent counterparts. Rule premises appear here a critical issue since they determine the structure of a rule base. This paper proposes a new approach to extracting fuzzy rules from training examples by means of genetic-based premise learning. In order to construct a 'parsimonious' fuzzy model with high generalization ability, general premise structure allowing incomplete compositions of input variables as well as OR connectives of linguistic terms is considered. A genetic algorithm is utilized to optimize both the premise structure of rules and fuzzy set membership functions at the same time. Determination of rule conclusions is nested in the premise learning, where consequences of individual rules are determined under fixed preconditions. The proposed method was applied to the well-known gas furnace data of Box and Jenkins to show its validity and to compare its performance with those of other works.     

Modeling Connectionist Neuro-Fuzzy network and Applications

The main objective of this paper is to propose a Neuro-Fuzzy network, which can model a system from input–output data by automatically dividing the input–output space and extracting fuzzy if-then rules from numerical data. The structure of the network is simple with input, rule and output layers only. The connections between input and rule layer is used to derive the membership functions of the fuzzified part. Kohonens self-organizing learning algorithm is applied to partition the pattern space. Using this algorithm, similar rules are mapped close by and extraction of if-then rules is made easy. It can also adapt to a number of rules automatically. The proposed network is verified for three benchmark applications. Experimental results show that the adaptive method discussed in this paper not only trains in a few hundred iterations but also provides better performance measures when compared to conventional methods.