Reliable All-Pairs Evolving Fuzzy Classifiers

Abstract

In this paper, we propose a novel design of evolving fuzzy classifiers (EFCs) to handle online multiclass classification problems in a data-streaming context. Therefore, we exploit the concept of all-pairs (AP), a.k.a. all-versus-all, classification using binary classifiers for each pair of classes. This benefits from less complex decision boundaries to be learned, as opposed to a direct multiclass approach, and achieves a higher efficiency in terms of incremental training time than one-versus-rest classification techniques. For the binary classifiers, we apply fuzzy classifiers with singleton class labels in the consequences, as well as Takagi-Sugeno (T-S) fuzzy models to conduct regression on [0, 1] for each class pair. Both are evolved and incrementally trained in a data-streaming context, yielding a permanent update of the whole AP collection of classifiers, thus being able to properly react to dynamic changes in the streams. The classification phase considers a novel strategy by using the preference levels of each pair of classes that are collected in a preference relation matrix and performing a weighted voting scheme on this matrix. This is done by investigating the reliability of the classifiers in their predictions: 1) integrating the degree of ignorance on samples to be classified as weights for the preference levels and 2) new conflict models used in the single binary classifiers and when calculating the final class response based on the preference relation matrix. The advantage of the new EFC concept over the single model (using a direct multiclass classification concept) and multimodel architectures (using a one-versus-rest classification concept) will be underlined by empirical evaluations and comparisons at the end of this paper based on high-dimensional real-world multiclass classification problems. The results also show that integrating conflict and ignorance concepts into the preference relations can boost classifier accuracies.

SO-ARTIST: Self-Organized ART-2A inspired clustering for online Takagi–Sugeno fuzzy models

“…These studies' measurement of the quality of rules after their merging or splitting is very appealing and an interesting point of view that can further enhance our work. Furthermore, recent research Lughofer and Buchtala [49] suggests that the usage of all-pairs architecture outperforms the multi-model architecture that we use in this work. Thus, we want to explore this modeling and possibly incorporate it into our solutions.…”

An efficient algorithm for large-scale quasi-supervised learning

“…The computational expense recorded the calculation time of the posterior probability estimates p 0 ðx i Þ and p 1 ðx i Þ for all samples x i including the minimization of EðnÞ for the optimal reference set size parameter n. The recognition accuracy was determined in terms of the detection and false alarm rates defined in Eqs. (20) and (21), and the area under the receiver operating characteristic curves generated by the P D -P FA graphs was computed.…”

“…Across the spectrum of statistical learning algorithms requiring various degrees of guidance from available data, supervised pattern classification algorithms such as the nearest neighbor rule [6,9], support vector machines [5,25], artificial neural networks [12], discriminant functions [21], and fuzzy classifiers [1,18,20] have enjoyed a particularly wide audience ranging from object recognition to biomedical data analysis. Such a far-reaching pertinence can be attributed to the ability of these algorithms to construct decision rules based on a given set of training samples for which the desired decisions are already available.…”

Rule-based OneClass-DS learning algorithm

“…A different approach, based on fuzzy classifiers [38][39][40], uses a fuzzy classifier learning algorithm to create rules from data with noise. For instance, fuzzy classifiers eClass and FLEXFIS-Class [41] can be used with different model architectures for solving one-class problems.…”