In class-incremental learning, the objective is to learn a number of classes sequentially without having access whole training data. However, due problem known as catastrophic forgetting, neural networks suffer substantial performance drop in such settings. The often approached by experience replay, method that stores limited samples be replayed future steps reduce forgetting learned classes. W...

This paper presents an investigation into the classification of a difficult data set containing large intra-class variability but low inter-class variability. Standard classifiers are weak and fail to achieve satisfactory results however, it is proposed that a combination of such weak classifiers can improve overall performance. The paper also introduces a novel evolutionary approach to fuzzy r...

Image-based approaches based on one-class classifiers are presented. The information is extracted with a feature-based representation and recognized by using an ensemble of one-class classifiers. The features extracted by ‘‘FingerCode’’ are used to capture the ridge strength. The experiments show that our system outperforms the standard ‘‘FingerCode’’ recognition method. r 2006 Elsevier B.V. Al...

We consider learning decision trees in the boosting framework, where we assume that the classifiers in each internal node come from a hypothesis class HI which satisfies the weak learning assumption. In this work we consider the class of stochastic linear classifiers for HI , and derive efficient algorithms for minimizing the Gini index for this class, although the problem is non-convex. This i...

This paper presents an ensemble of neo-cognitron neural network base classifiers to enhance the accuracy of the system, along the experimental results. The method offers lesser computational preprocessing in comparison to other ensemble techniques as it ex-preempts feature extraction process before feeding the data into base classifiers. This is achieved by the basic nature of neo-cognitron, it...

In this article we analyze the effect of class distribution on classifier learning. We begin by describing the different ways in which class distribution affects learning and how it affects the evaluation of learned classifiers. We then present the results of two comprehensive experimental studies. The first study compares the performance of classifiers generated from unbalanced data sets with ...

Class imbalance learning is an important research problem in data mining and machine learning. Most solutions including levels, algorithm cost sensitive approaches are derived using multi-class classifiers, depending on the number of classes to be classified. One-class classification (OCC) techniques, contrast, have been widely used for anomaly or outlier detection where only normal positive cl...

Several researchers have proposed effective approaches for binary classification in the last years. We can easily extend some of those techniques to multi-class. Notwithstanding, some other powerful classifiers (e.g., SVMs) are hard to extend to multi-class. In such cases, the usual approach is to reduce the multi-class problem complexity into simpler binary classification problems (divide-and-...

Lung nodule detection is a class imbalanced problem because nodules are found with much lower frequency than nonnodules. In the class imbalanced problem, conventional classifiers tend to be overwhelmed by the majority class and ignore the minority class. We therefore propose cascaded convolutional neural networks to cope with the class imbalanced problem. In the proposed approach, cascaded conv...