Returns the Quantile in the range [0.0 .. 1.0] for an numerically typed attributes. The evaluation is performed on the value that is present on this event compared to the full population (past events) within the configured Time Window.
Unlike Statistic Of Numeric, which can be used to detect anomalies in behaviour of an individual user or device, Quantile Of Numeric allows detection of anomalies with respect to the full user population.
For Example:
- Comparing the Amount of a Purchase or Payment against past events to see if it unusually large or small compared to the full population without having to define arbitrary thresholds (which require constant monitoring and updating).
- Detecting outlier devices by comparing Device or TCP attributes such as profiling.tcp_connection[*].rtt or profiling.tcp_connection[*].mss
- Detecting outlier behaviour by comparing Journey or Biometric attributes such as time_since_last_step
Configuration
-
Feature Name
The "dictionary" name under which the feature value will be stored. -
Numeric Attribute
- The attribute to calculate the quantile of. Only attributes that have a Real or Integer data type can be used.
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.
Use Case 1
Implementing In The Feature Editor
- name: purchase_amt_quantile
default_value: 0
time_window:
days: 7
find_the:
quantile_of_numeric:
attribute: purchase.amount
with_scope: same_node_instance