Fuzzy integral-based perceptron for two-class pattern classification problems

The single-layer perceptron with single output node is a well-known neural network for two-class classification problems. Furthermore, the sigmoid or logistic function is usually used as the activation function in the output neuron. A critical step is to compute the sum of the products of the connection weights with the corresponding inputs, which indicates the assumption of additivity among individual variables. Unfortunately, because the input variables are not always independent of each other, an assumption of additivity may not be reasonable enough. In this paper, the inner product can be replaced with an aggregation value obtained by a useful fuzzy integral by viewing each of the connection weights as a value of a λ-fuzzy measure for the corresponding variable. A genetic algorithm is then employed to obtain connection weights by maximizing the number of correctly classified training patterns and minimizing the errors between the actual and desired outputs of individual training patterns. The experimental results further demonstrate that the proposed method outperforms the traditional single-layer perceptron and performs well in comparison with other fuzzy or non-fuzzy classification methods.     

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.     

面向多类别模式分类问题的新型阴性选择算法

针对阴性选择算法缺乏高效的分类器生成机制和“过拟合”抑制机制的缺陷,提出了一种面向多类别模式分类的阴性选择算法CS-NSA。通过引入克隆选择机制,根据分类器的分类效果和刺激度对其进行自适应学习;针对多类别模式分类的“过拟合”问题,引入了检测器集合的修剪机制,增强了检测器的分类推广能力。对比实验结果证明：与著名的人工免疫分类器AIRS相比,CS-NSA体现出更高的正确识别率。     

不平衡分类问题研究综述

实际的分类问题往往都是不平衡分类问题,采用传统的分类方法,难以得到满意的分类效果。为此,十多年来,人们相继提出了各种解决方案。对国内外不平衡分类问题的研究做了比较详细地综述,讨论了数据不平衡性引发的问题,介绍了目前几种主要的解决方案。通过仿真实验,比较了具有代表性的重采样法、代价敏感学习、训练集划分以及分类器集成在3个实际的不平衡数据集上的分类性能,发现训练集划分和分类器集成方法能较好地处理不平衡数据集,给出了针对不平衡分类问题的分类器评测指标和将来的工作。     

Analytic network process for pattern classification problems using genetic algorithms

The analytic network process (ANP) is a useful technique for multi-attribute decision analysis (MCDA) that employs a network representation to describe interrelationships between diverse attributes. Owing to effectiveness of the ANP in allowing for complex interrelationships between attributes, this paper develops an ANP-based classifier for pattern classification problems with interdependence or independence among attributes. To deal with interdependence, this study employs genetic algorithms (GAs) to automatically determine elements in the supermatrix that are not easily user-specified, to find degrees of importance of respective attributes. Then, with the relative importance for each attribute in the limiting supermatrix, the current work determines the class label of a pattern by its synthetic evaluation. Experimental results obtained by the proposed ANP-based classifier are comparable to those obtained by other fuzzy or non-fuzzy classification methods.     

Texture pattern analysis of kidney tissues for disorder identification and classification using dominant Gabor wavelet

A study on ultrasound kidney images using proposed dominant Gabor wavelet is made for classifying a few important kidney categories. Three kidney categories, namely, normal (NR), medical renal diseases (MRD) and cortical cyst (CC) are considered for the analysis. Of the 30 Gabor wavelets, a unique dominant Gabor wavelet is determined by maximizing the similarity between original pre-processed image and reconstructed Gabor image. The dominant Gabor features “m_mn^D{\mu_{mn}^D } ” and “AAD_mn^D{AAD_{mn}^D } ” are then evaluated to characterize the tissues of kidney region and compared with the Gabor features derived by considering all Gabor wavelets individually and as a whole using the resultant classification efficiency. The results obtained show that the proposed dominant Gabor wavelet features provide the classification efficiency of 86.66% for NR, 76.66% for MRD and 83.33% for CC, while individual wavelet features offer less than 70%, 63.33% and 66% for NR, MRD and CC. The overall classification efficiency improves by 18.89% with dominant Gabor features when compared to the classification efficiency obtained by considering all the Gabor wavelets features. The outputs of the proposed technique are validated with medical experts to assess the actual efficiency. The overall discriminating ability of the systems is also evaluated with performance evaluation measures, F-score and ROC. It has been observed that the dominant Gabor wavelet improves the classification efficiency appreciably and explores the possibility of implementing a computer-aided diagnosis system exclusively for ultrasound kidney images.     

Vectorial approximations of infinite-dimensional covariance descriptors for image classification

Computational Visual Media -     

Performance evaluation of fuzzy classifier systems formultidimensional pattern classification problems

We examine the performance of a fuzzy genetics-based machine learning method for multidimensional pattern classification problems with continuous attributes. In our method, each fuzzy if-then rule is handled as an individual, and a fitness value is assigned to each rule. Thus, our method can be viewed as a classifier system. In this paper, we first describe fuzzy if-then rules and fuzzy reasoning for pattern classification problems. Then we explain a genetics-based machine learning method that automatically generates fuzzy if-then rules for pattern classification problems from numerical data. Because our method uses linguistic values with fixed membership functions as antecedent fuzzy sets, a linguistic interpretation of each fuzzy if-then rule is easily obtained. The fixed membership functions also lead to a simple implementation of our method as a computer program. The simplicity of implementation and the linguistic interpretation of the generated fuzzy if-then rules are the main characteristic features of our method. The performance of our method is evaluated by computer simulations on some well-known test problems. While our method involves no tuning mechanism of membership functions, it works very well in comparison with other classification methods such as nonfuzzy machine learning techniques and neural networks.     

Self-generating prototypes for pattern classification

Prototype classifiers are a type of pattern classifiers, whereby a number of prototypes are designed for each class so as they act as representatives of the patterns of the class. Prototype classifiers are considered among the simplest and best performers in classification problems. However, they need careful positioning of prototypes to capture the distribution of each class region and/or to define the class boundaries. Standard methods, such as learning vector quantization (LVQ), are sensitive to the initial choice of the number and the locations of the prototypes and the learning rate. In this article, a new prototype classification method is proposed, namely self-generating prototypes (SGP). The main advantage of this method is that both the number of prototypes and their locations are learned from the training set without much human intervention. The proposed method is compared with other prototype classifiers such as LVQ, self-generating neural tree (SGNT) and K-nearest neighbor (K-NN) as well as Gaussian mixture model (GMM) classifiers. In our experiments, SGP achieved the best performance in many measures of performance, such as training speed, and test or classification speed. Concerning number of prototypes, and test classification accuracy, it was considerably better than the other methods, but about equal on average to the GMM classifiers. We also implemented the SGP method on the well-known STATLOG benchmark, and it beat all other 21 methods (prototype methods and non-prototype methods) in classification accuracy.     

Filter-based models for pattern classification

In this paper we consider a technique for pattern classification based upon the development of prototypes which capture the distinguishing features (“disjunctive prototypes”) of each pattern class and, via cross-correlation with incoming test images, enable efficient pattern classification. We evaluate such a classification procedure with prototypes based on the images per se (direct code), Gabor scheme (multiple fixed filter representation) and an edge (scale space-based) coding scheme. Our analyses, and comparisons with human pattern classification performance, indicate that the edge-only disjunctive prototypes provide the most discriminating classification performance and are the more representative of human behaviour.     

K-winner machines for pattern classification

The paper describes the K-winner machine (KWM) model for classification. KWM training uses unsupervised vector quantization and subsequent calibration to label data-space partitions. A K-winner classifier seeks the largest set of best-matching prototypes agreeing on a test pattern, and provides a local-level measure of confidence. A theoretical analysis characterizes the growth function of a K-winner classifier, and the result leads to tight bounds to generalization performance. The method proves suitable for high-dimensional multiclass problems with large amounts of data. Experimental results on both a synthetic and a real domain (NIST handwritten numerals) confirm the approach effectiveness and the consistency of the theoretical framework.     

A novel fuzzy classifier with Choquet integral-based grey relational analysis for pattern classification problems

This paper aims to propose a fuzzy classifier, which is a one-class-in-one-network structure consisting of multiple novel single-layer perceptrons. Since the output value of each single-layer perceptron can be interpreted as the overall grade of the relationship between the input pattern and one class, the degree of relationship between an attribute of the input pattern and that of this class can be taken into account by establishing a representative pattern for each class. A feature of this paper is that it employs the grey relational analysis to compute the grades of relationship for individual attributes. In particular, instead of using the sigmoid function as the activation function, a non-additive technique, the Choquet integral, is used as an activation function to synthesize the performance values, since an assumption of noninteraction among attributes may not be reasonable. Thus, a single-layer perceptron in the proposed structure performs the synthetic evaluation of the Choquet integral-based grey relational analysis for a pattern. Each connection weight is interpreted as a degree of importance of an attribute and can be determined by a genetic algorithm-based method. The experimental results further demonstrate that the test results of the proposed fuzzy classifier are better than or comparable to those of other fuzzy or non-fuzzy classification methods.     

Incorporation of user preference into multi-objective genetic fuzzy rule selection for pattern classification problems

In the design of fuzzy-rule-based systems, we have two conflicting objectives: accuracy maximization and interpretability maximization. As a measure of interpretability, a number of criteria have been proposed in the literature. Most of those criteria have been incorporated into fitness functions in order to automatically find accurate and interpretable fuzzy systems by genetic algorithms. However, interpretability is very subjective and is rarely defined for any users beforehand. In this article, we propose the incorporation of user preference into multi-objective genetic fuzzy rule selection for pattern classification problems. User preference is represented by a preference function which is changeable according to the user’s direct manipulation during evolution. The preference function is used as one of the objective functions in multi-objective genetic fuzzy rule selection. The effectiveness of the proposed method is examined through some case studies for the design of fuzzyrule-based classifiers.     

Texture classification using wavelet transform

Today, texture analysis plays an important role in many tasks, ranging from remote sensing to medical imaging and query by content in large image data bases. The main difficulty of texture analysis in the past was the lack of adequate tools to characterize different scales of textures effectively. The development in multi-resolution analysis such as Gabor and wavelet transform help to overcome this difficulty. This paper describes the texture classification using (i) wavelet statistical features, (ii) wavelet co-occurrence features and (iii) a combination of wavelet statistical features and co-occurrence features of one level wavelet transformed images with different feature databases. It is found that, the results of later method are promising.