Two-parameter Fisher criterion

Malina, Witold

doi:10.1109/3477.938265

Cited by 25 publications

(7 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2008.2000395 whether a classifier is used to evaluate the performance of the new feature set, these DDR techniques can be categorized into wrapper or filter methods, respectively. Feature extraction methods, e.g., principal component analysis (PCA) [10] and linear discriminant analysis (LDA) [38], [39], [46], transform the original set of features into a new set of features. Because the new features are different from the original features, it may be difficult to interpret the meaning of the new features.…”

Section: Introductionmentioning

confidence: 99%

A General Wrapper Approach to Selection of Class-Dependent Features

Wang

Zhou

Feng

2008

IEEE Trans. Neural Netw.

108

View full text Add to dashboard Cite

Abstract-In this paper, we argue that for a C-class classification problem, C 2-class classifiers, each of which discriminating one class from the other classes and having a characteristic input feature subset, should in general outperform, or at least match the performance of, a C-class classifier with one single input feature subset. For each class, we select a desirable feature subset, which leads to the lowest classification error rate for this class using a classifier for a given feature subset search algorithm. To fairly compare all models, we propose a weight method for the class-dependent classifier, i.e., assigning a weight to each model's output before the comparison is carried out. The method's performance is evaluated on two artificial data sets and several real-world benchmark data sets, with the support vector machine (SVM) as the classifier, and with the RELIEF, class separability, and minimal-redundancy-maximal-relevancy (mRMR) as attribute importance measures. Our results indicate that the class-dependent feature subsets found by our approach can effectively remove irrelevant or redundant features, while maintaining or improving (sometimes substantially) the classification accuracy, in comparison with other feature selection methods.Index Terms-Class-dependent feature extraction, class-dependent feature selection, feature importance ranking, minimal-redundancy-maximal-relevancy (mRMR), one-against-all, one-against-one, support vector machine (SVM).

show abstract

Section: Introductionmentioning

confidence: 99%

A General Wrapper Approach to Selection of Class-Dependent Features

Wang

Zhou

Feng

2008

IEEE Trans. Neural Netw.

108

View full text Add to dashboard Cite

show abstract

“…From our experience and previous theoretical considerations (Malina, ), we know that the simple generalization of Fisher criterion from L =2 classes to multi‐class problems does not lead, in most of the cases, to the expected discriminant space; we think that such operation does not make sense. The detailed discussion of this topic along with the definition of the two‐parameter Fisher criterion can be found in Malina ().…”

Section: Methodsmentioning

confidence: 97%

“…The idea to test different combinations of matrices used in function was motivated by our experiences related to the search for the optimal form of Fisher criterion (Malina, ). The new variants of the SDF that we propose are listed in Table .…”

Section: Multi‐axis Sdf Algorithmsmentioning

confidence: 99%

SDF classifier revisited

Smiatacz

Malina

2011

Expert Systems

View full text Add to dashboard Cite

This paper is addressing problems related to the construction of classifiers based on the Similarity Discriminant Function (SDF), in which the traditional vector representation of a pattern is replaced with matrix data. We introduce potential modifications of the matrix data structure and propose new variants of the SDF. The algorithms that we present were tested on images of handwritten digits and on photographs of human faces, taken from the ORL and CMU-PIE databases. The results of experiments show that our modifications significantly improved the performance of the original SDF classifier.

show abstract

“…Zhang et al present a dual Eigenspace method for face recognition [15]. In [16] and [17], several new discrimination principles based on the Fisher criterion were proposed. Yang use kernel principal component analysis (PCA) for facial feature extraction and recognition [18], while Bartlett et al apply the independent component analysis (ICA) in face recognition [19].…”

Section: Introductionmentioning

confidence: 99%