Module: Column
Normalization
|
|
| Lesson Objective | Learn how to make data columns comparable for similarity and clustering calculations |
| Section | Description |
| Data Normalization | A discussion of data normalization and the default setting for viewing normalized data in a Profile chart. |
| Rationale for Creating Normalized Columns of Data | A discussion of the need to generate normalized data columns for computational analysis involving comparison of multivariate data profiles. |
| Creating Normalized Data Columns | How to generate normalized data columns using one of the following algorithms: Z-score, scale between 0 and 1, or divide by standard deviation. |
| Replacing Empty Values | A discussion of options for replacing empty values in the resulting data columns with a constant value, row average, column average, or interpolated value going down a column or across a row. |
| Column Normalization – Results | Examining the results of the column normalization process. |
Print