Heuristics Miner

This node implements the Heuristics Miner to discover a process model. Heuristics Miner is an algorithm that acts on the directly-follows Graph, providing way to handle with noise and to find dependency between activities. An heuristics net or causal net is the output, which is a directed graph with activities as nodes, dependency interpreted as arcs and bindings. The Heuristics Net can be then converted into a Petri net.
This node shows the Heuristics Net as one internal result in View "Heuristic Net", while it outputs a Petri net after an implicit conversion.


Event classifer
The event classifier is chosen to classify the event log.
All tasks connected
Every task needs to have at least one input and output arc, except the initial and the final activity.
Long distance dependency
Show long distance relations in the model.
Threshold: Relative-to-best
The admissible distance between directly follows relations for an activity and the activity's best one. At 0 only the best directly follows relation will be shown for every activity, at 100 all will be shown.
Threshold: Dependency
The strength of the directly follows relations determines when to show arcs (based on how frequently one activity is followed by another).
Threshold: Length-one-loops
Show arcs based on frequency of L1L observations
Threshold: Length-two-loops
Show arcs based on frequency of L2L observations
Threshold: Long distance
Show arcs based on how frequently one activity is eventually followed by another

Input Ports

An event log as input

Output Ports

The discovered process model in Petri net

Popular Predecessors

Popular Successors


Heuristic Net
Heuristic Net Visualization




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.