Data point importance: Information ranking in multivariate data

Zade, Somaye Vali; Neymeyr, Klaus; Sawall, Mathias; Fischer, Christian; Abdollahi, Hamid

doi:10.1002/cem.3453

“…However, this can be further optimised by, for example, limiting the number of convex peels or applying a threshold on the sample relevance criterion to compress the data more adequately, reducing computation times. Furthermore, other relevance criteria can be applied as well (see the recent work done by Zade et al 17 ).…”

Section: Discussionmentioning

confidence: 99%

Weighted multivariate curve resolution—Alternating least squares based on sample relevance

Ahmad

¹

,

Vitale

²

,

Cocchi

³

et al. 2023

Journal of Chemometrics

View full text Add to dashboard Cite

Alternating least squares within the multivariate curve resolution framework has seen a lot of practical applications and shows their distinction with their relatively simple and flexible implementation. However, the limitations of least squares should be carefully considered when deviating from the standard assumed data structure. Within this work, we highlight the effects of noise in the presence of minor components, and we propose a novel weighting scheme within the weighted multivariate curve-resolution-alternating least squares framework to resolve it. Two simulated and one Raman imaging case are investigated by comparing the novel methodology against standard multivariate curve resolution-alternating least squares and essential spectral pixel selection. A trade-off is observed between current methods, whereas the novel weighting scheme demonstrates a balance where the benefits of the previous two methods are retained.

show abstract

“…However, this can be further optimized by e.g., limiting the number of convex peels or applying a threshold on the sample relevance criterion to compress the data more adequately, reducing computation times. Furthermore, other relevance criteria can be applied as well (see the recent work done by Zade et al [17]).…”

Section: Discussionmentioning

confidence: 99%

Weighted multivariate curve resolution – alternating least squares based on sample relevance

Ahmad

¹

,

Vitale

²

,

Cocchi

³

et al. 2023

Preprint

0

View full text Add to dashboard Cite

Alternating least squares, within the multivariate curve resolution framework has seen a lot of practical applications and shows its distinction with its relatively simple and flexible implementation. However, the limitations of least squares should be considered carefully when deviating from the standard assumed data structure. Within this work we highlight the effects of noise in the presence of minor components, and we propose a novel weighting scheme within the weighted multivariate curve-resolution-alternating least squares framework, to resolve it. Two simulated and one Raman imaging case is investigated, by comparing the novel methodology against standard multivariate curve resolution-alternating least squares and essential spectral pixel selection. A trade-off is observed between current methods, while the novel weighting scheme demonstrates a balance, where the benefits of the previous two methods are retained.

show abstract

“…We propose an extended weighting scheme within the weighted MCR-ALS framework that is based on convex hull data peeling and is able to preserve the benefits of ESP selection without reducing model parameter stability. This weighting framework is based on samples relevance towards the MCR-ALS resolution, however, can be further optimized by e.g., applying a threshold on the sample relevance criterium to compress the data more adequately, reducing computation times, or using other relevance criteria (see the recent work done by Zade et al [16]).…”

Section: Discussionmentioning

confidence: 99%

Weighted multivariate curve resolution – alternating least squares based on sample relevance

Ahmad

¹

,

Vitale

²

,

Cocchi

³

et al. 2022

Preprint

1

0

View full text Add to dashboard Cite

Alternating least squares, within the multivariate curve resolution framework has seen a lot of practical applications and shows its distinction with its relatively simple and flexible implementation. However, the limitations of least squares should be considered carefully when deviating from the standard assumed data structure. Within this work we highlight the effects of noise in the presence of minor components, and we propose a novel weighting scheme within the weighted multivariate curve-resolution-alternating least squares framework, to resolve it. Two simulated and one Raman imaging case is investigated, by comparing the novel methodology against standard multivariate curve resolution-alternating least squares and essential spectral pixel selection. A trade-off is observed between current methods, while the novel weighting scheme demonstrates a balance, where the benefits of the previous two methods are retained.

show abstract

Data point importance: Information ranking in multivariate data

Cited by 7 publications

References 34 publications

Weighted multivariate curve resolution—Alternating least squares based on sample relevance

Weighted multivariate curve resolution—Alternating least squares based on sample relevance

Weighted multivariate curve resolution – alternating least squares based on sample relevance

Weighted multivariate curve resolution – alternating least squares based on sample relevance

Contact Info

Product

Resources

About