Overview
Manual de-identification is the configuration method in which each field rule is defined directly in the UI. A dataset is selected, a data type is added, and the De-identification option binds a generator, mode, and field path to a field.
Manual is one of three methods. The shared configuration sequence and the common reference tables (generators, modes, data types, output) are in De-identification Overview; this page covers what is specific to the manual method.
When to use it
- Defining rules field by field with full control.
- De-identifying a known set of fields without a template or AI detection.
Manual rule definition
| Element | Description |
| Dataset list | Datasets in the selected data source, listed on the left; one is selected to de-identify. |
| Add Data Type | Adds a data type to the selected dataset (reached via Edit). |
| Data Type | The field’s type (Bool, Datetime, Double, Float, Int, Long, String, Timestamp). |
| De-identification option | Chosen from the type’s dropdown; enables Generator, Mode, and Field Path. |
| Generator | The replacement generator. Full list in De-identification Overview. |
| Mode | Consistent, Random, or Pass Through. Definitions in De-identification Overview. |
| Field Path | Chosen from a picker that walks the dataset structure (parent then field). |
| Tag | Optional tag applied to the field. |
Key behaviors
Settings are gated by the De-identification option. Generator, Mode, and Field Path appear only after De-identification is chosen on the data type.
Field path is selected from the structure. The field path is chosen from a picker that walks the dataset (parent then field), binding the rule to a specific path in the document.
Multiple rules per configuration. Several rules can be defined before continuing to the filter step.Same rule model as Excel. The Generator, Mode, and Field Path chosen here correspond to the DEType, DEMethod, and Field Path columns in the Excel template.