de.unihd.dbs.uima.annotator.heideltime
public class HeidelTime extends org.apache.uima.analysis_component.JCasAnnotator_ImplBase
Modifier and Type | Field and Description |
---|---|
int |
timex_counter |
int |
timex_counter_global |
Constructor and Description |
---|
HeidelTime() |
Modifier and Type | Method and Description |
---|---|
void |
addTimexAnnotation(java.lang.String timexType,
int begin,
int end,
Sentence sentence,
java.lang.String timexValue,
java.lang.String timexQuant,
java.lang.String timexFreq,
java.lang.String timexMod,
java.lang.String timexId,
java.lang.String foundByRule,
org.apache.uima.jcas.JCas jcas)
Add timex annotation to CAS object.
|
java.lang.String |
applyRuleFunctions(java.lang.String tonormalize,
java.util.regex.MatchResult m) |
boolean |
checkPosConstraint(Sentence s,
java.lang.String posConstraint,
java.util.regex.MatchResult m,
org.apache.uima.jcas.JCas jcas)
Check whether the part of speech constraint defined in a rule is satisfied.
|
static java.lang.String |
correctDurationValue(java.lang.String value)
Durations of a finer granularity are mapped to a coarser one if possible, e.g., "PT24H" -> "P1D".
|
void |
deleteOverlappedTimexes(org.apache.uima.jcas.JCas jcas) |
void |
findTimexes(java.lang.String timexType,
java.util.HashMap<java.util.regex.Pattern,java.lang.String> hmPattern,
java.util.HashMap<java.lang.String,java.lang.String> hmOffset,
java.util.HashMap<java.lang.String,java.lang.String> hmNormalization,
java.util.HashMap<java.lang.String,java.lang.String> hmQuant,
Sentence s,
org.apache.uima.jcas.JCas jcas)
Apply the extraction rules, normalization rules
|
java.lang.String[] |
getAttributesForTimexFromFile(java.lang.String rule,
java.util.HashMap<java.lang.String,java.lang.String> hmNormalization,
java.util.HashMap<java.lang.String,java.lang.String> hmQuant,
java.util.HashMap<java.lang.String,java.lang.String> hmFreq,
java.util.HashMap<java.lang.String,java.lang.String> hmMod,
java.util.regex.MatchResult m,
org.apache.uima.jcas.JCas jcas) |
java.lang.String |
getPosFromMatchResult(int tokBegin,
int tokEnd,
Sentence s,
org.apache.uima.jcas.JCas jcas)
Identify the part of speech (POS) of a MarchResult.
|
void |
initialize(org.apache.uima.UimaContext aContext) |
void |
process(org.apache.uima.jcas.JCas jcas) |
void |
removeInvalids(org.apache.uima.jcas.JCas jcas)
Postprocessing: Remove invalid timex expressions.
|
void |
specifyAmbiguousValues(org.apache.uima.jcas.JCas jcas)
Under-specified values are disambiguated here.
|
getRequiredCasInterface, process
getCasInstancesRequired, hasNext, next
public int timex_counter
public int timex_counter_global
public void initialize(org.apache.uima.UimaContext aContext) throws org.apache.uima.resource.ResourceInitializationException
initialize
in interface org.apache.uima.analysis_component.AnalysisComponent
initialize
in class org.apache.uima.analysis_component.AnalysisComponent_ImplBase
org.apache.uima.resource.ResourceInitializationException
AnalysisComponent#initialize(UimaContext)
public void process(org.apache.uima.jcas.JCas jcas)
process
in class org.apache.uima.analysis_component.JCasAnnotator_ImplBase
JCasAnnotator_ImplBase.process(JCas)
public void addTimexAnnotation(java.lang.String timexType, int begin, int end, Sentence sentence, java.lang.String timexValue, java.lang.String timexQuant, java.lang.String timexFreq, java.lang.String timexMod, java.lang.String timexId, java.lang.String foundByRule, org.apache.uima.jcas.JCas jcas)
timexType
- begin
- end
- timexValue
- timexId
- foundByRule
- jcas
- public void removeInvalids(org.apache.uima.jcas.JCas jcas)
jcas
- public void specifyAmbiguousValues(org.apache.uima.jcas.JCas jcas)
jcas
- public void deleteOverlappedTimexes(org.apache.uima.jcas.JCas jcas)
jcas
- public java.lang.String getPosFromMatchResult(int tokBegin, int tokEnd, Sentence s, org.apache.uima.jcas.JCas jcas)
tokBegin
- tokEnd
- s
- jcas
- public void findTimexes(java.lang.String timexType, java.util.HashMap<java.util.regex.Pattern,java.lang.String> hmPattern, java.util.HashMap<java.lang.String,java.lang.String> hmOffset, java.util.HashMap<java.lang.String,java.lang.String> hmNormalization, java.util.HashMap<java.lang.String,java.lang.String> hmQuant, Sentence s, org.apache.uima.jcas.JCas jcas)
timexType
- hmPattern
- hmOffset
- hmNormalization
- hmQuant
- s
- jcas
- public boolean checkPosConstraint(Sentence s, java.lang.String posConstraint, java.util.regex.MatchResult m, org.apache.uima.jcas.JCas jcas)
s
- posConstraint
- m
- jcas
- public java.lang.String applyRuleFunctions(java.lang.String tonormalize, java.util.regex.MatchResult m)
public java.lang.String[] getAttributesForTimexFromFile(java.lang.String rule, java.util.HashMap<java.lang.String,java.lang.String> hmNormalization, java.util.HashMap<java.lang.String,java.lang.String> hmQuant, java.util.HashMap<java.lang.String,java.lang.String> hmFreq, java.util.HashMap<java.lang.String,java.lang.String> hmMod, java.util.regex.MatchResult m, org.apache.uima.jcas.JCas jcas)
public static java.lang.String correctDurationValue(java.lang.String value)
value
-