A Method of Dimensionality Reduction of Analog Circuit Fault Feature

We show that the full multiscale Persson's theory for rubber friction due to viscoelastic losses can be approximated extremely closely to simpler models, like that suggested by Persson in 1998 and similarly by Popov in his 2010 book (but notice that we do not make any use of the so-called “Method of Dimensionality Reduction” (MDR)), so it is essentially a single scale model at the so-called large wavevector cutoff. The dependence on the entire spectrum of roughness is therefore only confusing, at least for range of fractal dimensions of interest D≃2.2, and we confirm this with actual exact calculations and reference to recent data of Lorenz et al. Moreover, we discuss the critical assumption of the choice of the “free parameter” best fit truncation cutoff.

Download Full-text

A method of dimensionality reduction by selection of components in principal component analysis for text classification

Filomat ◽

10.2298/fil1805499z ◽

2018 ◽

Vol 32 (5) ◽

pp. 1499-1506 ◽

Cited By ~ 1

Author(s):

Yangwu Zhang ◽

Guohe Li ◽

Heng Zong

Keyword(s):

Principal Component Analysis ◽

Feature Extraction ◽

Dimensionality Reduction ◽

Principal Components Analysis ◽

Text Classification ◽

Principal Components ◽

Principal Component ◽

Method Of Dimensionality Reduction ◽

Components Analysis ◽

Selection Of

Dimensionality reduction, including feature extraction and selection, is one of the key points for text classification. In this paper, we propose a mixed method of dimensionality reduction constructed by principal components analysis and the selection of components. Principal components analysis is a method of feature extraction. Not all of the components in principal component analysis contribute to classification, because PCA objective is not a form of discriminant analysis (see, e.g. Jolliffe, 2002). In this context, we present a function of components selection, which returns the useful components for classification by the indicators of the performances on the different subsets of the components. Compared to traditional methods of feature selection, SVM classifiers trained on selected components show improved classification performance and a reduction in computational overhead.

Download Full-text

Text Categorisation Through Dimensionality Reduction Using Wavelet Transform

Journal of Information & Knowledge Management ◽

10.1142/s0219649220500392 ◽

2020 ◽

Vol 19 (04) ◽

pp. 2050039

Author(s):

Jorge Chamorro-Padial ◽

Rosa Rodríguez-Sánchez

Keyword(s):

Wavelet Transform ◽

Dimensionality Reduction ◽

Text Classification ◽

High Energy ◽

Discrete Wavelet ◽

Wavelet Coefficients ◽

Vertical Orientation ◽

Method Of Dimensionality Reduction ◽

Horizontal Orientation ◽

Relevant Level

This paper proposes a new method of dimensionality reduction when performing Text Classification, by applying the discrete wavelet transform to the document-term frequencies matrix. We analyse the features provided by the wavelet coefficients from the different orientations: (1) The high energy coefficients in the horizontal orientation correspond to relevant terms in a single document. (2) The high energy coefficients in the vertical orientation correspond to relevant terms for a single document, but not for the others. (3) The high energy coefficients in the diagonal orientation correspond to relevant terms in a document in comparison to other terms. If we filter using the wavelet coefficients and fulfil these three conditions simultaneously, we can obtain a reduced vocabulary of the corpus, with less dimensions than in the original one. To test the success of the reduced vocabulary, we recoded the corpus with the new reduced vocabulary and we obtained a statistically relevant level of accuracy for document classification.

Download Full-text

Impact of an elastic sphere with an elastic half space revisited: Numerical analysis based on the method of dimensionality reduction

Scientific Reports ◽

10.1038/srep08479 ◽

2015 ◽

Vol 5 (1) ◽

Cited By ~ 9

Author(s):

I. A. Lyashenko ◽

V. L. Popov

Keyword(s):

Numerical Analysis ◽

Dimensionality Reduction ◽

Half Space ◽

Elastic Half Space ◽

Elastic Sphere ◽

Method Of Dimensionality Reduction

Download Full-text

SIMULATION OF FRICTIONAL DISSIPATION UNDER BIAXIAL TANGENTIAL LOADING WITH THE METHOD OF DIMENSIONALITY REDUCTION

Facta Universitatis Series Mechanical Engineering ◽

10.22190/fume170503007d ◽

2017 ◽

Vol 15 (2) ◽

pp. 295

Author(s):

Andrey V. Dimaki ◽

Roman Pohrt ◽

Valentin L. Popov

Keyword(s):

Dimensionality Reduction ◽

Normal Load ◽

Superposition Principle ◽

Three Dimensional ◽

Method Of Dimensionality Reduction ◽

Elastic Bodies ◽

Frictional Dissipation ◽

Order Of Magnitude ◽

Constant Normal Load ◽

Numeric Solution

The paper is concerned with the contact between the elastic bodies subjected to a constant normal load and a varying tangential loading in two directions of the contact plane. For uni-axial in-plane loading, the Cattaneo-Mindlin superposition principle can be applied even if the normal load is not constant but varies as well. However, this is generally not the case if the contact is periodically loaded in two perpendicular in-plane directions. The applicability of the Cattaneo-Mindlin superposition principle guarantees the applicability of the method of dimensionality reduction (MDR) which in the case of a uni-axial in-plane loading has the same accuracy as the Cattaneo-Mindlin theory. In the present paper we investigate whether it is possible to generalize the procedure used in the MDR for bi-axial in-plane loading. By comparison of the MDR-results with a complete three-dimensional numeric solution, we arrive at the conclusion that the exact mapping is not possible. However, the inaccuracy of the MDR solution is on the same order of magnitude as the inaccuracy of the Cattaneo-Mindlin theory itself. This means that the MDR can be also used as a good approximation for bi-axial in-plane loading.

Download Full-text