Package beamline.models.algorithms

Class InfiniteSizeDirectlyFollowsMapper

java.lang.Object

org.apache.flink.api.common.functions.AbstractRichFunction

org.apache.flink.api.common.functions.RichFlatMapFunction<BEvent,T>

beamline.models.algorithms.StreamMiningAlgorithm<DirectlyFollowsRelation>

beamline.models.algorithms.InfiniteSizeDirectlyFollowsMapper

All Implemented Interfaces:

Serializable, org.apache.flink.api.common.functions.FlatMapFunction<BEvent,DirectlyFollowsRelation>, org.apache.flink.api.common.functions.Function, org.apache.flink.api.common.functions.RichFunction

public class InfiniteSizeDirectlyFollowsMapper
extends StreamMiningAlgorithm<DirectlyFollowsRelation>

This mapper transforms a stream of BEvents into a stream of DirectlyFollowsRelations. It transforms each pair of consequent events appearing in the same case as a directly follows operator (generating an object with type DirectlyFollowsRelation).

This mapper is called infinite because it's memory footprint will grow as the number of case ids grows as well.

Author:

Andrea Burattin

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor	Description
InfiniteSizeDirectlyFollowsMapper()

Method Summary

Modifier and Type	Method	Description
DirectlyFollowsRelation	ingest(BEvent event)	This abstract method is what each derive class is expected to implement.

Methods inherited from class beamline.models.algorithms.StreamMiningAlgorithm

flatMap, getProcessedEvents, open, process

Methods inherited from class org.apache.flink.api.common.functions.AbstractRichFunction

close, getIterationRuntimeContext, getRuntimeContext, setRuntimeContext

Methods inherited from class java.lang.Object

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

Constructor Detail

InfiniteSizeDirectlyFollowsMapper

public InfiniteSizeDirectlyFollowsMapper()

Method Detail

ingest

public DirectlyFollowsRelation ingest(BEvent event)

Description copied from class: StreamMiningAlgorithm

This abstract method is what each derive class is expected to implement. The argument of the method is the new observation and the returned value is the result of the mining.

If this method returns value null, then the value is not moved forward into the pipeline.

Specified by:

ingest in class StreamMiningAlgorithm<DirectlyFollowsRelation>

Parameters:

event - the new event being observed

Returns:

the result of the mining of the event, or null if nothing should go through the rest of the pipeline