Multi-class linear dimension reduction by generalized Fisher criteria
نویسندگان
چکیده
Linear Disciminant Analysis is in general unable to find the lower-dimensional feature space which maximizes the class discrimination, even if the class distributions can be assumed to be very simple, e.g. Gaussians with identical covariance matrices. In this paper we reformulate the K-class Fisher criterion as a sum of K(K 1)=2 2-class Fisher criteria. This formulation allows to weigh class pair contributions according to their relevance for classification. Further it offers an obvious way how to cope with heteroscedastic models. We propose a particular weighting scheme which attempts to approximate the pairwise Bayes error. Moderate improvements are obtained on the TIMIT phoneme classification task.
منابع مشابه
Multiclass Linear Dimension Reduction by Weighted Pairwise Fisher Criteria
ÐWe derive a class of computationally inexpensive linear dimension reduction criteria by introducing a weighted variant of the well-known K-class Fisher criterion associated with linear discriminant analysis (LDA). It can be seen that LDA weights contributions of individual class pairs according to the Euclidian distance of the respective class means. We generalize upon LDA by introducing a dif...
متن کاملA Multi Linear Discriminant Analysis Method Using a Subtraction Criteria
Linear dimension reduction has been used in different application such as image processing and pattern recognition. All these data folds the original data to vectors and project them to an small dimensions. But in some applications such we may face with data that are not vectors such as image data. Folding the multidimensional data to vectors causes curse of dimensionality and mixed the differe...
متن کاملLocal likelihood regression in generalized linear single-index models with applications to microarray data
Searching for an effective dimension reduction space is an important problem in regression, especially for high dimensional data such as microarray data. A major characteristic of microarray data consists in the small number of observations n and a very large number of genes p. This “large p, small n” paradigm makes the discriminant analysis for classification difficult. In order to offset this...
متن کاملManipulating Multi-qudit Entanglement Witnesses by Using Linear Programming
A new class of entanglement witnesses (EWs) called reduction type entanglement witnesses is introduced, which can detect some multi-qudit entangeled states including PPT ones with Hilbert space of dimension d1 ⊗ d2 ⊗ ... ⊗ dn . The novelty of this work comes from the fact that the feasible regions turn out to be convex polygons, hence the manipulation of these EWs reduces to linear programming ...
متن کاملAn intelligent fault diagnosis method of rolling bearings based on regularized kernel Marginal Fisher analysis
Generally, the vibration signals of fault bearings are non-stationary and highly nonlinear under complicated operating conditions. Thus, it’s a big challenge to extract optimal features for improving classification and simultaneously decreasing feature dimension. Kernel Marginal Fisher analysis (KMFA) is a novel supervised manifold learning algorithm for feature extraction and dimensionality re...
متن کامل