Data point importance: Information ranking in multivariate data. (26th October 2022)

• Record Type: Journal Article
• Authors: Zade, Somaye Vali; Neymeyr, Klaus; Sawall, Mathias; Fischer, Christian; Abdollahi, Hamid
• Is Part Of: Journal of chemometrics. Volume 37:Number 1(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-10-26
• Subjects: big data analysis -- data point importance -- data reduction -- hyperspectral image data -- information ranking -- pattern recognition data -- rotational ambiguity; Chemistry -- Mathematics -- Periodicals; Chemistry -- Statistical methods -- Periodicals; 542.85
• DOI: 10.1002/cem.3453
• Languages: English
• ISSNs: 0886-9383

Abstract

Abstract: A new characteristic term, data point importance (DPI), is introduced in order to quantify the information ranking in multivariate data. DPI defines an easily calculable value corresponding to each row or column of data matrix to reflect its impact for keeping the pattern of the data structure. Usually, a lot of data points have DPIs equal or very close to zero so that they do not carry on useful information about keeping the data pattern. DPI values for some of the data points are significant, and they have been sorted regarding to their importance values. The applicability of the proposed procedure is tested for three typical examples of simulated and experimental data including hyperspectral image data. The possible incorporation of the DPI values for more informative score plots is suggested.