GoldEntityAndRelationReader

Overview

Package

Class

Use

Tree

Deprecated

Index

Help

PREV CLASS NEXT CLASS

FRAMES NO FRAMES

SUMMARY: NESTED | FIELD | CONSTR | METHOD

DETAIL: FIELD | CONSTR | METHOD

org.apache.ctakes.relationextractor.cr
Class GoldEntityAndRelationReader

java.lang.Object
  org.apache.uima.analysis_component.AnalysisComponent_ImplBase
      org.apache.uima.analysis_component.Annotator_ImplBase
          org.apache.uima.analysis_component.JCasAnnotator_ImplBase
              org.uimafit.component.JCasAnnotator_ImplBase
                  org.apache.ctakes.relationextractor.cr.GoldEntityAndRelationReader

All Implemented Interfaces:: org.apache.uima.analysis_component.AnalysisComponent

public class GoldEntityAndRelationReader
extends org.uimafit.component.JCasAnnotator_ImplBase
extends org.uimafit.component.JCasAnnotator_ImplBase

Read named entity annotations and relations between them from knowtator xml files into the CAS. Assumptions: - A pair of entities can only have a single relation between them - An entity can have only a single semantic type For each relation instance in the gold standard, this reader will: - Check if the arguments of this relation instance can be extracted by CTAKEs automatically. If one of them cannot, this relation instance and the entities will be skipped. - Check if another relation between a pair of entities with the same knowtator mention ids has already been added to the cas. If it has, the reader will not add a new relation between these entities. This reader will also make sure each entity is added to the cas only once. E.g. the cas may already contain an entity if it participates in another relation that's already been added to the cas or due to an error in the gold standard (i.e. if it was annotated twice -- such weirdness does happen). TODO: Currently this reader does not normalize the roles of the arguments accross different corpora. It will simply add to the cas whatever is in the data. However, the roles were not consistently annotated accross different corpora (e.g. Sharp and Share assign different roles to the modifiers and entity mentions that participate in degree_of relation). This issue needs to be addressed so that modles can be trained on data coming from different sources.

Author:: dmitriy dligach

Field Summary
`int`	`identifiedAnnotationId`
`static java.io.File`	`inputDirectory`
`static java.lang.String`	`PARAM_INPUTDIR`
`int`	`relationArgumentId`
`int`	`relationId`

Constructor Summary
`GoldEntityAndRelationReader()`

Method Summary
`void`	`initialize(org.apache.uima.UimaContext aContext)`
`void`	`process(org.apache.uima.jcas.JCas jCas)`

Methods inherited from class org.uimafit.component.JCasAnnotator_ImplBase
`getLogger`

Methods inherited from class org.apache.uima.analysis_component.JCasAnnotator_ImplBase
`getRequiredCasInterface, process`

Methods inherited from class org.apache.uima.analysis_component.Annotator_ImplBase
`getCasInstancesRequired, hasNext, next`

Methods inherited from class org.apache.uima.analysis_component.AnalysisComponent_ImplBase
`batchProcessComplete, collectionProcessComplete, destroy, getContext, getResultSpecification, reconfigure, setResultSpecification`

Methods inherited from class java.lang.Object
`clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait`

Field Detail

PARAM_INPUTDIR

public static final java.lang.String PARAM_INPUTDIR

See Also:: Constant Field Values

inputDirectory

public static java.io.File inputDirectory

identifiedAnnotationId

public int identifiedAnnotationId

relationId

public int relationId

relationArgumentId

public int relationArgumentId

Constructor Detail