Parsimonious multiscale classification models

Alsberg, Bjørn K.

doi:10.1002/1099-128x(200009/12)14:5/6<529::aid-cem629>3.0.co;2-e

Cited by 14 publications

(4 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In general, almost all the feature selection methods described in the previous sections can be applied to WT coefficients; most used are filter and wrapper methods, e.g., GA [46,78]. Specific of feature selection in WT domain is that it can be applied scale/level-wise [79] or coefficients-wise [46,80,81], as well as the fact that WT decomposition parameters have to be set, such as a wavelet filter, decomposition level; some algorithms try to automatically iterate on these settings [46].…”

Section: Feature Selection In Wavelet Domainmentioning

confidence: 99%

Chemometric Methods for Classification and Feature Selection

Cocchi

Biancolillo

Marini

2018

Comprehensive Analytical Chemistry

View full text Add to dashboard Cite

Section: Feature Selection In Wavelet Domainmentioning

confidence: 99%

Chemometric Methods for Classification and Feature Selection

Cocchi

Biancolillo

Marini

2018

Comprehensive Analytical Chemistry

View full text Add to dashboard Cite

“…The wavelet transform is a popular signal-processing technique for use in data compression and feature extraction. It has been successfully applied to various kinds of analyses of chemical process data. ,− Further, variable selection has previously been applied to wavelet coefficients to extract information from raw data and to build parsimonious empirical models for spectrum calibration and classification. , In our study, the wavelet transform is used to extract frequency- and time-specific information from batch trajectories of process variables by using a novel two-stage variable-selection strategy.…”

Section: Wavelet Transformmentioning

confidence: 99%

“…1,[16][17][18][19][20][21] Further, variable selection has previously been applied to wavelet coefficients to extract information from raw data and to build parsimonious empirical models for spectrum calibration and classification. 22,23 In our study, the wavelet transform is used to extract frequency-and time-specific information from batch trajectories of process variables by using a novel two-stage variable-selection strategy.…”

Section: Overview Of the Wavelet Transformmentioning

confidence: 99%

Two-Stage Variable Selection Using the Wavelet Transform of Batch Trajectories for Data Interpretation and Construction of Parsimonious Quality-Estimation Models

Chu

Kim

Han

et al. 2007

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

We propose a two-stage variable-selection strategy performed in the wavelet domain in order to extract quality-related information from batch trajectories of process variables and to build parsimonious quality-estimation models. The proposed variable-selection method proceeds in two stages and uses the discrete wavelet transform of the batch trajectories. This approach greatly reduces the computation time required for finding those wavelet coefficients related to product quality. A quality-estimation model built with the selected wavelet coefficient subset is shown for the case study we discuss to exhibit satisfactory estimation accuracy. The application of the method to an industrial PVC polymerization process was found to achieve the same prediction accuracy as a multiway partial least-squares (MPLS) based model that uses all variables in the time domain.

show abstract

“…Also, a spectral dataset often comprises more variables/features than samples. There is an increased interest in producing models that have a minimum of variables without losing the prediction ability [13]. Thus, variable/feature selection strategies are commonly used before building classification/regression models.…”

Section: Introductionmentioning

confidence: 99%