Random Classify Log

This node classifies each trace variant from the event log according to the given label and its distribution. Trace variant is a set of traces which have the same event classes sequence. The following inputs are required: a given label, a list of possible values and their corresponding distributions. The sum of distributions are 1.0. Otherwise, it gives one configuration exception.
For example, to classify one event log by attribute "kpi" with the following distribution, {"pos", 0.5} and {"neg", 0.5}. The configuration steps are :

  1. Given the Label Name "kpi" at the top of the table;
  2. Click the button "Add label value" to add value for this label and its percent.
  3. Input value "pos" to Label values and 0.5 to Label value percentages.
  4. Input value "neg" to Label values and 0.5 to Label value percentages.
After execution, given 10 traces for 2 trace variants in the event log, then for each trace variant, 5 traces are assigned with "pos" for label "kpi", 5 traces are assigned as "neg". If the sum doesn't match the total trace number, the one with the highest percent will be assigned to the difference value.

Options

Label Name
The category name for classification
Label Values
The possible label value.
Label Value Percentage
The percentage of corresponding label value for one trace variant. The value is between 0 - 1.0. The sum of the all percentage values should be 1.

Input Ports

Icon
Event Log to classify by randomly

Output Ports

Icon
Event log with labels after classification

Popular Predecessors

  • No recommendations found

Popular Successors

  • IncorporateNegInfo100 %

Views

This node has no views

Workflows

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.