/**
* Converts NamedEntityTagAnnotation tags into {@link EntityMention}s. This finds the longest
* sequence of NamedEntityTagAnnotation tags of the matching type.
*
* @param sentence A sentence annotated with NamedEntityTagAnnotation
*/
public void makeAnnotationFromAllNERTags(CoreMap sentence) {
List<CoreLabel> words = sentence.get(CoreAnnotations.TokensAnnotation.class);
List<EntityMention> mentions =
sentence.get(MachineReadingAnnotations.EntityMentionsAnnotation.class);
assert words != null;
if (mentions == null) {
this.logger.info("mentions are null");
mentions = new ArrayList<>();
}
for (int start = 0; start < words.size(); start++) {
int end;
// find the first token after start that isn't of nerType
String lastneTag = null;
String ne = null;
for (end = start; end < words.size(); end++) {
ne = words.get(end).get(NamedEntityTagAnnotation.class);
if (ne.equals(SeqClassifierFlags.DEFAULT_BACKGROUND_SYMBOL)
|| (lastneTag != null && !ne.equals(lastneTag))) {
break;
}
lastneTag = ne;
}
if (end > start) {
// found a match!
String entityType = this.getEntityTypeForTag(lastneTag);
EntityMention m =
EntityMentionFactory.constructEntityMention(
EntityMention.makeUniqueId(),
sentence,
new Span(start, end),
new Span(start, end),
entityType,
null,
null);
// TODO: changed entityType in the above sentence to nerTag - Sonal
logger.info("Created " + entityType + " entity mention: " + m);
start = end - 1;
mentions.add(m);
}
}
sentence.set(MachineReadingAnnotations.EntityMentionsAnnotation.class, mentions);
}
public EntityMention makeEntityMention(
CoreMap sentence, int start, int end, String label, String identifier) {
Span span = new Span(start, end);
String type = null, subtype = null;
if (!label.startsWith("B-") && !label.startsWith("I-")) {
type = label;
subtype = null; // TODO: add support for subtypes! (needed at least in ACE)
} else {
type = label.substring(2);
subtype = null; // TODO: add support for subtypes! (needed at least in ACE)
}
EntityMention entity =
EntityMentionFactory.constructEntityMention(
identifier, sentence, span, span, type, subtype, null);
Counter<String> probs = new DefaultCounter<>();
probs.setCount(entity.getType(), 1.0);
entity.setTypeProbabilities(probs);
return entity;
}
/**
* Converts NamedEntityTagAnnotation tags into {@link EntityMention}s. This finds the longest
* sequence of NamedEntityTagAnnotation tags of the matching type.
*
* @param sentence A sentence, ideally annotated with NamedEntityTagAnnotation
* @param nerTag The name of the NER tag to copy, e.g. "DATE".
* @param entityType The type of the {@link EntityMention} objects created
*/
public static void makeAnnotationFromGivenNERTag(
CoreMap sentence, String nerTag, String entityType) {
List<CoreLabel> words = sentence.get(CoreAnnotations.TokensAnnotation.class);
List<EntityMention> mentions =
sentence.get(MachineReadingAnnotations.EntityMentionsAnnotation.class);
assert words != null;
assert mentions != null;
for (int start = 0; start < words.size(); start++) {
int end;
// find the first token after start that isn't of nerType
for (end = start; end < words.size(); end++) {
String ne = words.get(end).get(NamedEntityTagAnnotation.class);
if (!ne.equals(nerTag)) {
break;
}
}
if (end > start) {
// found a match!
EntityMention m =
EntityMentionFactory.constructEntityMention(
EntityMention.makeUniqueId(),
sentence,
new Span(start, end),
new Span(start, end),
entityType,
null,
null);
logger.info("Created " + entityType + " entity mention: " + m);
start = end - 1;
mentions.add(m);
}
}
sentence.set(MachineReadingAnnotations.EntityMentionsAnnotation.class, mentions);
}
