com.aliasi.coref.MentionChainImpl Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of aliasi-lingpipe Show documentation
Show all versions of aliasi-lingpipe Show documentation
This is the original Lingpipe:
http://alias-i.com/lingpipe/web/download.html
There were not made any changes to the source code.
/*
* LingPipe v. 4.1.0
* Copyright (C) 2003-2011 Alias-i
*
* This program is licensed under the Alias-i Royalty Free License
* Version 1 WITHOUT ANY WARRANTY, without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the Alias-i
* Royalty Free License Version 1 for more details.
*
* You should have received a copy of the Alias-i Royalty Free License
* Version 1 along with this program; if not, visit
* http://alias-i.com/lingpipe/licenses/lingpipe-license-1.txt or contact
* Alias-i, Inc. at 181 North 11th Street, Suite 401, Brooklyn, NY 11211,
* +1 (718) 290-9170.
*/
package com.aliasi.coref;
import com.aliasi.coref.matchers.EntityTypeMatch;
import com.aliasi.coref.matchers.ExactPhraseMatch;
import com.aliasi.coref.matchers.GenderKiller;
import com.aliasi.coref.matchers.HonorificConflictKiller;
import com.aliasi.coref.matchers.SequenceSubstringMatch;
import com.aliasi.coref.matchers.SynonymMatch;
/**
* An implementation of mention chains.
*
* @author Bob Carpenter
* @version 3.9.1
* @since LingPipe1.0
*/
public class MentionChainImpl extends AbstractMentionChain {
/**
* Identifier for matching functions for this chain.
*/
private int mMatcherDescriptor;
/**
* Construct a mention chain implementation containing the
* single specified mention at the specified offset, assigning
* the chain the specified identifier.
*
* @param mention Mention to add to mention chain.
* @param offset Sentence offset of mention in document.
* @param id Identifier for resulting mention chain.
*/
public MentionChainImpl(Mention mention, int offset, int id) {
super(mention,offset,id);
mMatcherDescriptor = typeToMatcherDescriptor(mention.entityType());
}
/**
* The matching functions defined for this mention chain.
*
* @return The matching functiosn for this mention chain.
*/
@Override
public Matcher[] matchers() {
if (mMatcherDescriptor < 0) return MATCHERS[THING_MATCHERS];
return MATCHERS[mMatcherDescriptor];
}
/**
* The killing functions defined for this mention chain.
*
* @return The killing functions for this mention chain.
*/
@Override
public Killer[] killers() {
if (mMatcherDescriptor < 0) return KILLERS[THING_MATCHERS];
return KILLERS[mMatcherDescriptor];
}
/**
* Adds the specified mention appearing at the specified
* sentence offset to this mention chain. Computes the
* new set of matchers for this descriptor, as well as
* setting the gender if the specified mention resolves it.
*
* @param mention Mention to add to this chain.
*/
@Override
protected void add(Mention mention) {
int typeToMatchIndex = typeToMatcherDescriptor(mention.entityType());
if (mMatcherDescriptor >= 0 && typeToMatchIndex >= 0) {
mMatcherDescriptor
= UNIFIER[mMatcherDescriptor][typeToMatchIndex];
}
if (gender() == null && mention.gender() != null)
setGender(mention.gender());
for (String honorific : mention.honorifics())
addHonorific(honorific);
}
/**
* The mapping from entity types to matching function
* descriptors
*/
private static int typeToMatcherDescriptor(String entityType) {
// could replace with switch on first char
if (entityType.equals(Tags.PERSON_TAG))
return MALE_OR_FEMALE_MATCHERS;
if (entityType.equals(Tags.LOCATION_TAG))
return THING_MATCHERS;
if (entityType.equals(Tags.ORGANIZATION_TAG))
return THING_MATCHERS;
if (entityType.equals(Tags.MALE_PRONOUN_TAG))
return MALE_PRONOUN_MATCHERS;
if (entityType.equals(Tags.FEMALE_PRONOUN_TAG))
return FEMALE_PRONOUN_MATCHERS;
return -1;
}
/**
* Adds unifications for specified descriptor with itself,
* resulting in itself.
*
* @param descriptor Descriptor to add to the collection.
*/
private static final void descriptor(int descriptor) {
subsume(descriptor,descriptor);
}
/**
* Adds unification for a description that is subsumed by a
* another descriptor, with the unification being equal
* to the more specific descriptor.
*
* @param descriptorLess Less informative descriptor.
* @param descriptorMore More informative descriptor.
*/
private static void subsume(int descriptorLess, int descriptorMore) {
unify(descriptorLess,descriptorMore,descriptorMore);
}
/**
* Adds unification for specified descriptors yielding specified
* result.
*
* @param descriptor1 First descriptor to unify.
* @param descriptor2 Second descriptor to unify.
* @return Result of unifying the two specified descriptors.
*/
private static void unify(int descriptor1, int descriptor2, int result) {
UNIFIER[descriptor1][descriptor2] = result;
UNIFIER[descriptor2][descriptor1] = result;
}
/**
* Identifier for matchers of either males or females.
*/
private static final int MALE_OR_FEMALE_MATCHERS = 0; // e1f3m3s3u1w1
/**
* Identifier for matchers for things.
*/
private static final int THING_MATCHERS = 1; // e1s3u1w1
/**
* Identifier for matchers for just male pronouns.
*/
private static final int MALE_PRONOUN_MATCHERS = 2; // m1
/**
* Identifier for matchers for just female pronouns.
*/
private static final int FEMALE_PRONOUN_MATCHERS = 3; // f1
/**
* Identifier for matchers for male persons.
*/
private static final int MALE_MATCHERS = 4; // e1m1s3u1w1
/**
* Identifier for matchers for female persons.
*/
private static final int FEMALE_MATCHERS = 5; // e1f1s3u1w1
/**
* The table storing the results of unifying descriptors.
*/
private static final int[][] UNIFIER = new int[6][6];
static {
descriptor(MALE_OR_FEMALE_MATCHERS);
descriptor(THING_MATCHERS);
descriptor(MALE_PRONOUN_MATCHERS);
descriptor(FEMALE_PRONOUN_MATCHERS);
descriptor(MALE_MATCHERS);
descriptor(FEMALE_MATCHERS);
subsume(MALE_PRONOUN_MATCHERS,MALE_MATCHERS);
subsume(FEMALE_PRONOUN_MATCHERS,FEMALE_MATCHERS);
subsume(MALE_OR_FEMALE_MATCHERS,MALE_MATCHERS);
subsume(MALE_OR_FEMALE_MATCHERS,FEMALE_MATCHERS);
unify(MALE_OR_FEMALE_MATCHERS,MALE_PRONOUN_MATCHERS,
MALE_MATCHERS);
unify(MALE_OR_FEMALE_MATCHERS,FEMALE_PRONOUN_MATCHERS,
FEMALE_MATCHERS);
}
/**
* The array mapping matcher descriptors into arrays of
* matchers.
*/
private static final Matcher[][] MATCHERS = new Matcher[6][];
static {
MATCHERS[MALE_OR_FEMALE_MATCHERS]
= new Matcher[] {
new ExactPhraseMatch(1),
new EntityTypeMatch(3,Tags.MALE_PRONOUN_TAG),
new EntityTypeMatch(3,Tags.FEMALE_PRONOUN_TAG),
new SequenceSubstringMatch(3),
new SynonymMatch(1)
};
MATCHERS[THING_MATCHERS]
= new Matcher[] {
new ExactPhraseMatch(1),
new SequenceSubstringMatch(3),
new SynonymMatch(1)
};
MATCHERS[MALE_PRONOUN_MATCHERS]
= new Matcher[] {
new EntityTypeMatch(1,Tags.MALE_PRONOUN_TAG),
};
MATCHERS[FEMALE_PRONOUN_MATCHERS]
= new Matcher[] {
new EntityTypeMatch(1,Tags.FEMALE_PRONOUN_TAG),
};
MATCHERS[MALE_MATCHERS]
= new Matcher[] {
new ExactPhraseMatch(1),
new EntityTypeMatch(1,Tags.MALE_PRONOUN_TAG),
new SequenceSubstringMatch(3),
new SynonymMatch(1)
};
MATCHERS[FEMALE_MATCHERS]
= new Matcher[] {
new ExactPhraseMatch(1),
new EntityTypeMatch(1,Tags.FEMALE_PRONOUN_TAG),
new SequenceSubstringMatch(3),
new SynonymMatch(1)
};
}
/**
* The array mapping matcher descriptors into arrays of
* killers.
*/
private static final Killer[][] KILLERS = new Killer[6][];
static {
KILLERS[MALE_OR_FEMALE_MATCHERS]
= new Killer[] {
new GenderKiller(),
new HonorificConflictKiller()
};
KILLERS[THING_MATCHERS]
= new Killer[] {
new GenderKiller()
};
KILLERS[MALE_PRONOUN_MATCHERS]
= new Killer[] {
new GenderKiller()
};
KILLERS[FEMALE_PRONOUN_MATCHERS]
= new Killer[] {
new GenderKiller()
};
KILLERS[MALE_MATCHERS]
= new Killer[] {
new GenderKiller(),
new HonorificConflictKiller()
};
KILLERS[FEMALE_MATCHERS]
= new Killer[] {
new GenderKiller(),
new HonorificConflictKiller()
};
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy