Documentation Index

Fetch the complete documentation index at: https://docs.darwinium.com/llms.txt

Use this file to discover all available pages before exploring further.

Specificity Of

  • Published on Jun 7, 2024
  • 1 minute(s) read
  • CG
 Prev Next

Has two modes of operation that can be used to determine how unique one or more attribute is with respect to the general population.

image.png

In its first mode of operation (Normalize into Fraction of Total: Yes), Specificity Of can automatically compute the normalized specificity without needing to configure additional features (that contain the approx total count of and perform the divide operation to normalize the value). This is provided as a convenience function for cases where they underlying values are not required.

image.png

In its second mode of operation (Normalize into Fraction of Total: No), Specificity Of calculates a high accuracy estimation of the number of times that a group of Identifier/Subject attributes with the same value as the current event occured over the given time period. This figure can be divided into the Approx Total Count of the same attributes in order to calculate specificity of the attributes in the range 0..1 (therefore normalizing the value). This can be useful when both the approximate velocity, and the specificity are required to be used in other more complex models or calculations.

For Example:

  • Determining if the value of one or more attributes is an outlier.
  • Determining if the value of one or more attributes is normal.

Configuration

  • Feature Name
    The "dictionary" name under which the feature value will be stored.
  • Attributes - selects which attribute(s) to measure the unique value/combination(s) of
    • Any attribute or combination of attributes may be selected.
Optional Configuration
  • Default Value
    A value that can be assigned to the feature if the attributes to calculate the features are not present.
  • Scope
    Can be used to extend the event search from the local node, to all nodes within the same organization all the way to all across customers (global).
  • Time Window
    Can be used to limit the Feature calculation to a specific time period/window. e.g 1 week.
Note

There is no restriction on Identifier or Subject attributes - any attribute(s) may be selected for use in Specificity Of.

Use Case 1

Using Specificity Of to determine if the operating system detected by Javascript profiling is an outlier with respect to the general population.

Implementing In The Feature Editor

image.png

image.png

  - name: spec_js_os_month
    default_value: 0
    time_window:
      months: 1
    find_the:
      specificity_of:
        attributes:
          - profiling.javascript.os
        event_type: all
        normalize: true

Using within Rules

outcome['CHAMPION'].features.general['spec_js_os_month'] <= 0.01

image.png