Providing PRTools with fuzzy rule-based classifiers

Abstract

This paper first reviews the state-of-the-art of fuzzy rule-based classifiers (FRBCs), then it discusses how to implement an FRBC under the Pattern Recognition Toolbox (PRTools), the de-facto standard toolbox for classification in Matlab. Such an implementation, called frbc, allows for a straightforward comparison of frbc with other classifiers already available under the PRTools. Furthermore, frbc can easily be used and combined with any other general-purpose function already available in PRTools. In this way, e.g., it becomes really easy to perform many types of feature selection, based on the accuracy achieved by frbc on the subset of features at hand. Another useful feature is the capability to export each FRBC generated by frbc as a standard Fuzzy Inference System (FIS) structure used within the Matlab Fuzzy Logic Toolbox (FLT): this allows comparisons/validations, visual inspection of the rule base, etc. In the experimental part we first assess the correctness of the implementation, by reproducing results existing in the literature. Then we show some examples of usage of frbc, combined with existing PRTools functions

A hierarchical approach to multi-class fuzzy classifiers

“…The FRM available in frbc (Cococcioni et al, 2010) is a general model of fuzzy reasoning for combining information provided by different rules. It is an extension, presented in Cordón, del Jesus, and Herrera (1999), of the fuzzy classifier defined by Kuncheva (1996).…”

“…The FRBC used in this paper is frbc, presented in Cococcioni et al (2010) and developed under the Pattern Recognition Toolbox (PRTools) (Duin et al, 2004), the de facto standard toolbox for classification in Matlab Ò . frbc follows the PRTools base philosophy, e.g., use of fast and heuristic-based training algorithms, function reuse, powerful and concise syntax, automatic training from data, total compatibility with the Matlab Ò environment.…”

“…The training technique available in frbc (Cococcioni et al, 2010) is the Wang and Mendel method extended to classification problems, an adaptation of the well-known Wang and Mendel method for regression problems (Wang & Mendel, 1992). This method includes the following steps: STEP A: generate a uniform fuzzy partition of the input domain, made of Q f , f = 1,. .…”

“…The fuzzy system is obtained exploiting a hierarchical scheme, as a combination of fuzzy models built (employing the fuzzy rule-based classifier frbc (Cococcioni, D'Andrea, & Lazzerini, 2010)) on input domain regions increasingly smaller, according to a multi-level grid-like partition. Only the necessary partitions are built, in order to avoid the explosion of the number of rules with the increase of the hierarchical level.…”

HFRBC-GA: A fuzzy classifier for energy systems applications

“…The fuzzy system is obtained exploiting a hierarchical scheme, as a combination of fuzzy models built (employing the fuzzy rule-based classifier frbc [27]) on input domain regions increasingly smaller, according to a multi-level grid-like partition. Only the necessary partitions are built, in order to avoid the explosion of the number of rules with the increase of the hierarchical level.…”