org.processmining.hadoop.AlphaMiner.algorithms.utils

Class AlphaMinerNodeExpander

java.lang.Object

org.processmining.hadoop.AlphaMiner.algorithms.utils.AlphaMinerNodeExpander

All Implemented Interfaces:

org.processmining.framework.util.search.NodeExpander<Tuple<java.lang.String>>

public class AlphaMinerNodeExpander
extends java.lang.Object
implements org.processmining.framework.util.search.NodeExpander<Tuple<java.lang.String>>

AlphaMiner Node Expander. The implementation is adapted from the one in the Alpha Miner ProM plugin

Author:

shernandez

Constructor Summary

Constructors

Constructor and Description
AlphaMinerNodeExpander(HadoopGraph dfg)

Method Summary

Methods

Modifier and Type	Method and Description
java.util.Collection<Tuple<java.lang.String>>	expandNode(Tuple<java.lang.String> toExpand, org.processmining.framework.plugin.Progress progress, java.util.Collection<Tuple<java.lang.String>> resultsSoFar) This method is called by the searcher when a node in the search tree has to be considered for expansion.
void	processLeaf(Tuple<java.lang.String> toAdd, org.processmining.framework.plugin.Progress progress, java.util.Collection<Tuple<java.lang.String>> resultCollection) This method is called by the searcher to which this expander is registered each time a leaf was encountered.
protected java.util.Collection<org.processmining.framework.util.Pair<org.processmining.models.graphbased.directed.petrinet.elements.Transition,java.lang.String>>	reverse(java.util.Map<java.lang.String,org.processmining.models.graphbased.directed.petrinet.elements.Transition> class2transition) Flip the mapping given around (so values map to keys); handles multiple values with same key.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

AlphaMinerNodeExpander

public AlphaMinerNodeExpander(HadoopGraph dfg)

Method Detail

expandNode

public java.util.Collection<Tuple<java.lang.String>> expandNode(Tuple<java.lang.String> toExpand,
                                                       org.processmining.framework.plugin.Progress progress,
                                                       java.util.Collection<Tuple<java.lang.String>> resultsSoFar)

Description copied from interface: org.processmining.framework.util.search.NodeExpander

This method is called by the searcher when a node in the search tree has to be considered for expansion. The node to expand is given as a parameter and the method should return a collection of nodes representing the children of this node. If the node turns out to be a leaf, an empty collection has to be returned.

Specified by:

expandNode in interface org.processmining.framework.util.search.NodeExpander<Tuple<java.lang.String>>

Parameters:

toExpand - The node to expand.

progress - The progress indicator provided to the searcher in which this expander is registered. The expander may increment the progress, but it should check for cancellation, especially when doing long computations.

resultsSoFar - the leafs of the search space that have been added to the resultCollection by the processLeaf method. Note that any access to this collection should be synchronized and this collection should not be modified in this method. However the provided collection is modifiable, as for synchronization purposes, the actual resultCollection is provided.

Returns:

A collection of child nodes, or an empty collection if this node turns out to be a leaf.

processLeaf

public void processLeaf(Tuple<java.lang.String> toAdd,
               org.processmining.framework.plugin.Progress progress,
               java.util.Collection<Tuple<java.lang.String>> resultCollection)

Description copied from interface: org.processmining.framework.util.search.NodeExpander

This method is called by the searcher to which this expander is registered each time a leaf was encountered. The provided resultCollection was specified in the startSearch method of the searcher and can be used to store this result. A typical implementation is: synchronized(resultCollection) { resultCollection.add(leaf); } However, more advanced implementations are possible. Finally, note that calls to this method are not thread-safe, i.e. the implementing class should take care of synchronization on the resultCollection if necessary (as in the example above).

Specified by:

processLeaf in interface org.processmining.framework.util.search.NodeExpander<Tuple<java.lang.String>>

Parameters:

toAdd - The node that was found to be a leaf by the expandNode method (i.e. the expandNode method returned and empty collection)

progress - The progress indicator provided to the searcher in which this expander is registered. The expander may increment the progress, but it should check for cancellation, especially when doing long computations.

resultCollection - The collection to which to add the leaf node. More advanced computations are allowed here, i.e. the resultCollection can be changed. Note however that this requires syncrhonization on the collection, as calls to this method are not thread-safe.

reverse

protected java.util.Collection<org.processmining.framework.util.Pair<org.processmining.models.graphbased.directed.petrinet.elements.Transition,java.lang.String>> reverse(java.util.Map<java.lang.String,org.processmining.models.graphbased.directed.petrinet.elements.Transition> class2transition)

Flip the mapping given around (so values map to keys); handles multiple values with same key.

Parameters:

class2transition -

Returns: