Autoscaling
Term · Chemistry · MLC-T-CHM-001005
A data preprocessing technique in chemometrics that scales each variable to have a mean of zero and a standard deviation of one. This is achieved by first subtracting the mean of the variable from each data point (mean centering) and then dividing by the variable's standard deviation. The result is that all variables are given equal weight in subsequent multivariate analyses, such as principal component analysis.
| Identifier | MLC-T-CHM-001005 |
|---|---|
| Field | Chemistry |
| Subject | Analytical Chemistry |
| References | PAC, 2016, 88, 407. 'Vocabulary of concepts and terms in chemometrics (IUPAC Recommendations 2016)' on page 409 (https://doi.org/10.1515/pac-2015-0605) |
| See also | Variance scaling; Mean centering |
Record as JSON
{
"id": "MLC-T-CHM-001005",
"term": "Autoscaling",
"field": "Chemistry",
"definition": "A data preprocessing technique in chemometrics that scales each variable to have a mean of zero and a standard deviation of one. This is achieved by first subtracting the mean of the variable from each data point (mean centering) and then dividing by the variable's standard deviation. The result is that all variables are given equal weight in subsequent multivariate analyses, such as principal component analysis.",
"subject": "Analytical Chemistry",
"see_also": [
"variance scaling",
"mean centering"
],
"references": [
"PAC, 2016, 88, 407. 'Vocabulary of concepts and terms in chemometrics (IUPAC Recommendations 2016)' on page 409 (https://doi.org/10.1515/pac-2015-0605)"
],
"url": "https://mlchart.com/terminology/chemistry/autoscaling/"
}
Record 1,083 of 13,676 in Chemistry terminology (MLC-0109). Request the full dataset.