public static String spanToLemmas(Span<Term> span) { StringBuffer stringBuffer = new StringBuffer(); for (Term term : span.getTargets()) { stringBuffer.append(term.getLemma().toLowerCase()); stringBuffer.append(" "); } return stringBuffer.toString().trim(); }
public Multimap<Term, T> match(final KAFDocument document, final Iterable<Term> terms) { Preconditions.checkNotNull(document); final Set<Term> termSet = ImmutableSet.copyOf(terms); final Multimap<Term, T> result = HashMultimap.create(); for (final Term term : termSet) { final String lemma = term.getLemma(); final String stem = Stemming.stem(null, lemma); for (final T lexeme : ImmutableSet.copyOf(Iterables.concat( this.lemmaIndex.get(term.getLemma()), this.stemIndex.get(stem)))) { if (lexeme.match(document, termSet, term)) { result.put(term, lexeme); } } } return result; }
public static String extendedPos(final KAFDocument document, final Term term) { final String pos = term.getMorphofeat(); final String lemma = term.getLemma().toLowerCase(); if ("some".equals(lemma) || "many".equals(lemma) || "all".equals(lemma) || "few".equals(lemma) || "this".equals(lemma) || "these".equals(lemma) || "that".equals(lemma) || "those".equals(lemma)) { final Dep dep = document.getDepToTerm(term); if (dep == null || !"NMOD".equals(dep.getRfunc())) { return pos + "P"; // determiner (DT) or adj (JJ) used as demonstrative pronoun } } return pos; }
public boolean match(@Nullable final Term term) { return term != null && (this.pos == null || this.pos.equalsIgnoreCase(term.getPos()) || this.pos .equals(term.getMorphofeat())) && (this.lemma == null || this.lemma.equalsIgnoreCase(term.getLemma())) && (this.stem == null || this.stem.equalsIgnoreCase(Stemming.stem(null, term.getStr()))); }
if (!func.equalsIgnoreCase("COORD") && !func.equals("CONJ")) { builder.append(builder.length() > 0 ? "_" : "").append( term.getLemma().toLowerCase().replace(' ', '_'));
sb.append(thisTerm.getForm()); sb.append("\t"); sb.append(thisTerm.getLemma()); sb.append("\t"); sb.append(thisTerm.getMorphofeat()); sb.append(thisTerm.getForm()); sb.append("\t"); sb.append(thisTerm.getLemma()); sb.append("\t"); sb.append(thisTerm.getMorphofeat()); sb.append(thisTerm.getForm()); sb.append("\t"); sb.append(thisTerm.getLemma()); sb.append("\t"); sb.append(thisTerm.getMorphofeat());
sb.append(thisTerm.getForm()); sb.append("\t"); sb.append(thisTerm.getLemma()); sb.append("\t"); sb.append(thisTerm.getMorphofeat()); sb.append(thisTerm.getForm()); sb.append("\t"); sb.append(thisTerm.getLemma()); sb.append("\t"); sb.append(thisTerm.getMorphofeat()); sb.append(thisTerm.getForm()); sb.append("\t"); sb.append(thisTerm.getLemma()); sb.append("\t"); sb.append(thisTerm.getMorphofeat());
final String readableHeadSynsetID = WordNet.getReadableSynsetID(headSynsetID); final String headID = MoreObjects.firstNonNull(readableHeadSynsetID, // head.getLemma().toLowerCase());
if (term.getLemma().equals(lemma)) { mainTerm = term;
final String lemma = term.getLemma(); if (pos == 'V') { final List<PropBank.Roleset> rolesets = PropBank.getRolesets(lemma);
this.model.add(mention.uri, KS.LEMMA, this.vf.createLiteral(head.getLemma()));
for (final Term term : predicate.getSpan().getTargets()) { builder.append(builder.length() == 0 ? "" : "_"); builder.append(term.getLemma().toLowerCase());
emitMeta(mentionID, KS.LEMMA, head.getLemma());
if (!entity.isNamed() || isLinked) { entityURI = mintURI(entity.getId(), entity.isNamed() ? entity.getSpans().get(0) .getStr() : head.getLemma()); } else { entityURI = Statements.VALUE_FACTORY.createURI(Util.cleanIRI("entity:"
final String expectedLemma = head.getLemma();
final String lemma = term.getLemma().toLowerCase(); if (sumoRefs.isEmpty() && synsetRef != null && !lemma.equals("be")) { Set<String> synsetIDs = Sets.newHashSet(synsetRef.getReference());
predicate.getId(), head.getLemma()); ann.predicateURI = predicateURI;
termElem.setAttribute("lemma", term.getLemma());
termElem.setAttribute("lemma", term.getLemma());