A brief taxonomy and ranking of creative prototype reduction schemes

Abstract

Various Prototype Reduction Schemes (PRS) have been reported in the literature. Based on their operating characteristics, these schemes fall into two fairly distinct categories -those which are of a creative sort, and those which are essentially selective. The norms for evaluating these methods are typically, the reduction rate and the classification accuracy. It is generally believed that the former class of methods is superior to the latter.In this paper, we report the results of executing various creative PRS, and attempt to comparatively quantify their capabilities. The paper presents a brief taxonomy of the various reported PRS schemes. Our experimental results for three artificial data sets, and for samples involving real-life data sets, demonstrate that no single method is uniformly superior to the others for all kinds of applications. This result, though consistent with the findings of [2], is, in one sense, counter-intuitive, because the various researchers have presented their specific PRS with the hope that it would be superior to the previously reported methods. But the fact is that while one method is superior in certain domains, it is inferior to another method when dealing with a data set with markedly different characteristics. The conclusion of this study is that the question of determining when one method is superior to another remains open. Indeed, it appears as if the designers of the Pattern Recognition system will have to choose the appropriate PRS based to the specific characteristics of the data that they are studying. The paper also suggests answers to various hypotheses that relate to the accuracies and reduction rates of families of PRS.

Differential evolution for optimizing the positioning of prototypes in nearest neighbor classification

“…Concretely, HYB combines support vector machines with LVQ3 and executes a search in order to find the most appropriate parameters of LVQ3 [31].…”

“…Well known methods for PG are PNN [28], learning quantization vector (LVQ) [29], Chen's algorithm [30], ICPL [27], HYB [31] and MixtGauss [32]. A good study of PS and PG can be found in [33].…”

“…For example, some PG methods implement ENN or DROP algorithms as early filtering processes [27,54] and, in [31], a hybridization method based on LVQ3 post-processing of conventional prototype reduction approaches is proposed.…”

“…The second contribution of this paper follows a similar idea to that presented in [31], but it is extended so that PS methods can be hybridized with any positioning adjustment of prototype method. Specifically, we study the use of LVQ3, PSO and DE algorithms for the optimization positioning of prototypes after a PS process.…”

Ultimate Order Statistics-Based Prototype Reduction Schemes

“…From an overall perspective, we now discuss how we are to achieve our goal to reduce the cardinality of the OS-based PRS to be unity for each class. First of all, we know that PRSs can be broadly classified as being "selective" or "creative" [12]. A "selective" PRS yields as its output a set of prototypes which are chosen from the original training points.…”