List of usage examples for org.apache.commons.lang3 StringUtils isAllUpperCase
public static boolean isAllUpperCase(final CharSequence cs)
Checks if the CharSequence contains only uppercase characters.
null will return false .
From source file:edu.illinois.cs.cogcomp.wikifier.utils.spelling.SurfaceFormSpellChecker.java
public static String getCorrection(String text) { // All uppercase normalization HONG KONG => Hong Kong String noPunc = text.replaceAll("[^A-Z0-9]*", ""); if (StringUtils.isAllUpperCase(noPunc) && noPunc.length() > 3) { char[] letters = text.toLowerCase().toCharArray(); for (int i = 0; i < letters.length; i++) { if (i == 0 || !Character.isLetter(letters[i - 1]) && Character.isLetter(letters[i])) { letters[i] = Character.toUpperCase(letters[i]); }/* ww w.ja v a2s .c o m*/ } return new String(letters); } // Spell check if (correctionCache.get(text) != null) { return String.valueOf(correctionCache.get(text)); } else { if (caching) { String correction = getGoogleCorrection(text); correctionCache.put(text, correction); return correction; } } return text; }
From source file:de.micromata.tpsb.doc.TpsbEnvUtils.java
public static boolean isGenericType(String type) { return /* type.length() < 3 && */StringUtils.isAllUpperCase(type) == true; }
From source file:de.micromata.tpsb.doc.ParserContext.java
public String getFullQualifiedNameFromImports(String className) { if (StringUtils.isEmpty(className) == true) { return className; }//from ww w . j ava2s .c om // Generics enthalten? className = removeGenerics(className); // es ist ein generischer Typ if (StringUtils.isAllUpperCase(className)) { return className; } // Klassenname ist leer oder bereits fullqualified if (className.indexOf(".") > -1) { return className; } String importedClass = currentImports.getKey(className); if (StringUtils.isNotEmpty(importedClass)) { return importedClass; } return currentPackage + "." + className; }
From source file:FunctionParser.MathExp2SPARQL.java
public MathExp2SPARQL(String StringFunction, List<Provider> prov, List<Usage> usage) { Parser p = new Parser(true); Map<String, Operator> temp = p.getFactory().getIdentifier2OperatorMap(); Iterator<String> it = temp.keySet().iterator(); while (it.hasNext()) { String key = it.next().toString(); String val = temp.get(key).getOperatorString(); //System.out.println(key + " " +val); //print the key-value entry opmap.put(key, val); }/*from www.ja va 2 s. co m*/ if (StringFunction.contains("IF")) { SPARQLQuery = "SELECT ?result\n" + "WHERE {\n"; String[] partitions = StringFunction.split("~"); ArrayList<String> parcels = new ArrayList<String>(); int counter = 0; for (String part : partitions) { String[] data = part.split(";"); String cond = data[0].replace("IF", ""); cond = cond.replace("ELSEIF", ""); cond = cond.replace("ELSE", ""); String form = data[1]; //System.out.println("Cond: "+cond+"\nForm: "+form); stack.clear(); infixstack.clear(); usageVariables.clear(); ASTNode objcond = p.parse(cond); convert(objcond); String fa = prefixToInfix(stack, infixstack); fa = fa.replace("[", ""); fa = fa.replace("]", ""); for (String s : usageVariables)//here we can detect the type of the variable by matching the variables name with the JAVA Object. Through the JAVA object we can see if its a Qualitative or Quantitative Value and treat it as such { if (!SPARQLQuery.contains(s)) { if (!StringUtils.isAllUpperCase(s)) SPARQLQuery = SPARQLQuery + "\n:" + s + " price:hasValue ?" + s + "_instance .\n" + "?" + s + "_instance gr:hasValue ?" + s + "_value .\n"; else { //add syntax to deal with constant qualitative attributes like, WINDOWS will probably be 'windows' in the SPARQL Query } } } stack.clear(); infixstack.clear(); usageVariables.clear(); ASTNode objform = p.parse(form); convert(objform); String fb = prefixToInfix(stack, infixstack); fb = fb.replace("[", ""); fb = fb.replace("]", ""); for (String s : usageVariables)//here we can detect the type of the variable by matching the variables name with the JAVA Object. Through the JAVA object we can see if its a Qualitative or Quantitative Value and treat it as such { if (!SPARQLQuery.contains(s)) { if (!StringUtils.isAllUpperCase(s)) SPARQLQuery = SPARQLQuery + "\n:" + s + " price:hasValue ?" + s + "_instance .\n" + "?" + s + "_instance gr:hasValue ?" + s + "_value .\n"; else { //add syntax to deal with constant qualitative attributes like, WINDOWS will probably be 'windows' in the SPARQL Query } } } //BIND (IF(((?gbs > 1) && (?gbs <= (10 * 1024))), ((?gbs - 1) * ?price10), 0) AS ?priceA) . String parcel = "BIND(IF(( " + fa + " ),(" + fb + "),0) AS ?result" + counter++ + " )."; parcels.add(parcel); //System.out.println("Cond : "+fa+"\nForm: "+fb+"\n***************************"); } // BIND ((((?priceA + ?priceB) + ?priceC) + ?priceD) AS ?price) . String sum = ""; counter--; for (; counter >= 0; counter--) { if (counter == 0) sum = sum + "?result" + counter; else sum = sum + "?result" + counter + "+"; } String lastParcel = "BIND((" + sum + ") AS ?result ) ."; for (String s : parcels) SPARQLQuery = SPARQLQuery + s + "\n"; SPARQLQuery = SPARQLQuery + lastParcel + "\n"; SPARQLQuery = SPARQLQuery + "\n}\n"; //System.out.println(SPARQLQuery); } else { ASTNode obj = p.parse(StringFunction);//insert the mathematical formula here convert(obj); SPARQLQuery = "SELECT ?result\n" + "WHERE {\n"; for (String s : usageVariables)//here we can detect the type of the variable by matching the variables name with the JAVA Object. Through the JAVA object we can see if its a Qualitative or Quantitative Value and treat it as such { if (!StringUtils.isAllUpperCase(s)) SPARQLQuery = SPARQLQuery + "\n:" + s + " price:hasValue ?" + s + "_instance .\n" + "?" + s + "_instance gr:hasValue ?" + s + "_value .\n"; else { } //add syntax to deal with constant qualitative attributes like, WINDOWS will probably be 'windows' in the SPARQL Query } String f = prefixToInfix(stack, infixstack); f = f.replace("[", ""); f = f.replace("]", ""); SPARQLQuery = SPARQLQuery + "\nBIND((" + f + ") AS ?result ) .\n" + "}"; } }
From source file:Heuristics.TermLevelHeuristics.java
public boolean isFirstLetterCapitalized(String termOrig) { return (StringUtils.isAllUpperCase(StringUtils.left(termOrig, 1))) ? true : false; }
From source file:edu.toronto.cs.cidb.ncbieutils.NCBIEUtilsAccessService.java
private static String fixCase(String text) { if (text == null || text.length() == 0) { return ""; }/*www . j av a2s .c o m*/ if (StringUtils.isAllUpperCase(text.replaceAll("[^a-zA-Z]", ""))) { return StringUtils.capitalize(text.toLowerCase()); } return text; }
From source file:FunctionParser.MathExp2SPARQL.java
public IExpr convert(ASTNode node) throws ConversionException { if (node == null) { return null; }//ww w. j a va 2s . c o m if (node instanceof FunctionNode) { //function node final FunctionNode functionNode = (FunctionNode) node; final IAST ast = F.ast(convert((ASTNode) functionNode.get(0))); for (int i = 1; i < functionNode.size(); i++) { ast.add(convert((ASTNode) functionNode.get(i))); } // code below return ast; } if (node instanceof FractionNode) { //fraction node FractionNode fr = (FractionNode) node; //System.out.println(node.toString()); stack.add("(" + node.toString() + ")"); // if (fr.isSign()) { // return F.fraction((IInteger) convert(fr.getNumerator()),(IInteger) convert(fr.getDenominator())).negate(); //only need the fraction as a string // } // return F.fraction( // (IInteger) convert(((FractionNode) node).getNumerator()),(IInteger) convert(((FractionNode) node).getDenominator())); } if (node instanceof PatternNode) { //pattern node final PatternNode pn = (PatternNode) node; return F.pattern((ISymbol) convert(pn.getSymbol()), convert(pn.getConstraint())); } if (node instanceof SymbolNode) {//symbol node if (SYMBOLS_MAP.containsKey(node.getString().toLowerCase())) { //System.out.println("Operator - " + node.getString()); stack.add(opmap.get(node.toString())); } else { //System.out.println("Variable - " + node.getString()); if (StringUtils.isAllUpperCase(node.getString())) stack.add(node.getString()); else stack.add("?" + node.getString().concat("_value")); usageVariables.add(node.getString()); } return F.symbol(node.getString()); } if (node instanceof IntegerNode) { // integer node final IntegerNode integerNode = (IntegerNode) node; final String iStr = integerNode.getString(); if (iStr != null) { //System.out.println("IntegerNode - "+F.integer(iStr, integerNode.getNumberFormat())); stack.add("" + F.integer(iStr, integerNode.getNumberFormat())); return F.integer(iStr, integerNode.getNumberFormat()); } //System.out.println("IntegerNode - " + integerNode.getIntValue()); stack.add("" + integerNode.getIntValue()); return F.integer(integerNode.getIntValue()); } if (node instanceof StringNode) { //string node //System.out.println("StringNode - " + node.getString()); return F.stringx(node.getString()); } if (node instanceof FloatNode) { //float node //System.out.println("FloatNode - " + node.getString()); stack.add(node.getString()); return F.num(node.getString()); } return F.symbol(node.toString()); }
From source file:Heuristics.TermLevelHeuristics.java
public boolean isAllCaps(String termOrig) { String temp = termOrig.replaceAll(" ", "").trim(); return (StringUtils.isAllUpperCase(temp)) ? true : false; }
From source file:it.cnr.isti.hpc.dexter.disambiguation.TurkishEntityDisambiguator.java
@Override public EntityMatchList disambiguate(DexterLocalParams localParams, SpotMatchList sml) { entityScoreMap = new HashMap<String, EntityScores>(); selectedEntities = new HashSet<String>(); Multiset<String> entityFrequencyMultiset = HashMultiset.create(); EntityMatchList entities = sml.getEntities(); String inputText = localParams.getParams().get("text"); String algorithm = Property.getInstance().get("algorithm"); String ambigious = Property.getInstance().get("algorithm.ambigious"); List<Token> inputTokens = Zemberek.getInstance().disambiguateFindTokens(inputText, false, true); List<Double> documentVector = DescriptionEmbeddingAverage.getAverageVectorList(inputText); Multiset<String> inputTokensMultiset = HashMultiset.create(); for (Token token : inputTokens) { inputTokensMultiset.add(token.getMorphText()); }//from w ww . j a v a 2s .c o m Multiset<String> domainMultiset = HashMultiset.create(); Multiset<String> typeMultiset = HashMultiset.create(); HashMap<String, Double> entitySimMap = new HashMap<String, Double>(); // if (printCandidateEntities) { // printEntities(entities); // } HashSet<String> words = new HashSet<String>(); Multiset<String> leskWords = HashMultiset.create(); // first pass for finding number of types and domains for (int i = 0; i < entities.size(); i++) { EntityMatch em = entities.get(i); String id = em.getId(); if (!entityFrequencyMultiset.contains(id)) { entityFrequencyMultiset.add(id); Entity entity = em.getEntity(); words.add(entity.getShingle().getText()); String type = entity.getPage().getType(); if (type != null && type.length() > 0) { typeMultiset.add(type); } String domain = entity.getPage().getDomain(); if (domain != null && domain.length() > 0) { domainMultiset.add(domain); } String desc = entity.getPage().getDescription(); List<Token> tokens = Zemberek.getInstance().disambiguateFindTokens(desc, false, true); for (Token token : tokens) { leskWords.add(token.getMorphText()); } } else { entityFrequencyMultiset.add(id); } } int maxDomainCount = 0; for (String domain : Multisets.copyHighestCountFirst(domainMultiset).elementSet()) { maxDomainCount = domainMultiset.count(domain); break; } int maxTypeCount = 0; for (String type : Multisets.copyHighestCountFirst(typeMultiset).elementSet()) { maxTypeCount = typeMultiset.count(type); break; } double maxSuffixScore = 0, maxLeskScore = 0, maxSimpleLeskScore = 0, maxLinkScore = 0, maxHashInfoboxScore = 0, maxwordvecDescriptionLocalScore = 0, maxHashDescriptionScore = 0, maxPopularityScore = 0, maxWordvectorAverage = 0, maxWordvecLinksScore = 0; // second pass compute similarities between entities in a window int currentSpotIndex = -1; SpotMatch currentSpot = null; for (int i = 0; i < entities.size(); i++) { EntityMatch em = entities.get(i); SpotMatch spot = em.getSpot(); if (currentSpot == null || spot != currentSpot) { currentSpotIndex++; currentSpot = spot; } String id = em.getId(); Entity entity = entities.get(i).getEntity(); EntityPage page = entities.get(i).getEntity().getPage(); String domain = page.getDomain(); String type = page.getType(); Shingle shingle = entity.getShingle(); /* windowing algorithms stars */ int left = currentSpotIndex - window; int right = currentSpotIndex + window; if (left < 0) { right -= left; left = 0; } if (right > sml.size()) { left += (sml.size()) - right; right = sml.size(); if (left < 0) { left = 0; } } double linkScore = 0, hashInfoboxScore = 0, wordvecDescriptionLocalScore = 0, hashDescriptionScore = 0, wordvecLinksScore = 0; for (int j = left; j < right; j++) { SpotMatch sm2 = sml.get(j); EntityMatchList entities2 = sm2.getEntities(); for (EntityMatch em2 : entities2) { String id2 = em2.getId(); EntityPage page2 = em2.getEntity().getPage(); int counter = 0; if (!ambigious.equals("true")) { for (EntityMatch entityMatch : entities2) { if (entityMatch.getId().startsWith("w")) { counter++; } } } if ((ambigious.equals("true") || counter == 1) && em.getSpot() != em2.getSpot() && !id.equals(id2)) { // Link Similarity calculation starts double linkSim = 0; if (id.startsWith("w") && id2.startsWith("w")) { if (entitySimMap.containsKey("link" + id + id2)) { linkSim = entitySimMap.get("link" + id + id2); } else { HashSet<String> set1 = Sets.newHashSet(page.getLinks().split(" ")); HashSet<String> set2 = Sets.newHashSet(page2.getLinks().split(" ")); linkSim = JaccardCalculator.calculateSimilarity(set1, set2); entitySimMap.put("link" + id + id2, linkSim); } linkScore += linkSim; // Link Similarity calculation ends } // Entity embedding similarity calculation starts double eeSim = 0; if (id.startsWith("w") && id2.startsWith("w")) { if (entitySimMap.containsKey("ee" + id + id2)) { eeSim = entitySimMap.get("ee" + id + id2); } else { eeSim = EntityEmbeddingSimilarity.getInstance().getSimilarity(page, page2); entitySimMap.put("ee" + id + id2, eeSim); } hashInfoboxScore += eeSim; } double w2veclinksSim = 0; if (id.startsWith("w") && id2.startsWith("w")) { if (entitySimMap.containsKey("wl" + id + id2)) { w2veclinksSim = entitySimMap.get("wl" + id + id2); } else { w2veclinksSim = AveragePooling.getInstance().getSimilarity(page.getWord2vec(), page2.getWord2vec()); entitySimMap.put("wl" + id + id2, w2veclinksSim); } wordvecLinksScore += w2veclinksSim; } // Entity embedding similarity calculation ends // Description word2vec similarity calculation // starts double word2vecSim = 0; if (entitySimMap.containsKey("w2v" + id + id2)) { word2vecSim = entitySimMap.get("w2v" + id + id2); } else { word2vecSim = AveragePooling.getInstance().getSimilarity(page2.getDword2vec(), page.getDword2vec()); entitySimMap.put("w2v" + id + id2, word2vecSim); } wordvecDescriptionLocalScore += word2vecSim; // Description word2vec similarity calculation ends // Description autoencoder similarity calculation // starts double autoVecSim = 0; if (entitySimMap.containsKey("a2v" + id + id2)) { autoVecSim = entitySimMap.get("a2v" + id + id2); } else { autoVecSim = AveragePooling.getInstance().getSimilarity(page2.getDautoencoder(), page.getDautoencoder()); entitySimMap.put("a2v" + id + id2, autoVecSim); } hashDescriptionScore += autoVecSim; // Description autoencoder similarity calculation // ends } } } if (linkScore > maxLinkScore) { maxLinkScore = linkScore; } if (hashInfoboxScore > maxHashInfoboxScore) { maxHashInfoboxScore = hashInfoboxScore; } if (wordvecDescriptionLocalScore > maxwordvecDescriptionLocalScore) { maxwordvecDescriptionLocalScore = wordvecDescriptionLocalScore; } if (hashDescriptionScore > maxHashDescriptionScore) { maxHashDescriptionScore = hashDescriptionScore; } if (wordvecLinksScore > maxWordvecLinksScore) { maxWordvecLinksScore = wordvecLinksScore; } /* windowing algorithms ends */ double domainScore = 0; if (domainMultiset.size() > 0 && maxDomainCount > 1 && domainMultiset.count(domain) > 1) { domainScore = (double) domainMultiset.count(domain) / maxDomainCount; } double typeScore = 0; if (typeMultiset.size() > 0 && maxTypeCount > 1 && typeMultiset.count(type) > 1) { typeScore = (double) typeMultiset.count(type) / maxTypeCount; } if (typeBlackList.contains(type)) { typeScore /= 10; } double typeContentScore = 0; if (type.length() > 0 && StringUtils.containsIgnoreCase(words.toString(), type)) { typeContentScore = 1; } double typeClassifierScore = TypeClassifier.getInstance().predict(page, page.getTitle(), page.getType(), entity.getShingle().getSentence()); double wordvecDescriptionScore = AveragePooling.getInstance().getSimilarity(documentVector, page.getDword2vec()); if (wordvecDescriptionScore > maxWordvectorAverage) { maxWordvectorAverage = wordvecDescriptionScore; } double suffixScore = 0; if (type != null && type.length() > 0) { Set<String> suffixes = new HashSet<String>(); String t = entity.getTitle().toLowerCase(new Locale("tr", "TR")); for (int x = 0; x < entities.size(); x++) { EntityMatch e2 = entities.get(x); if (e2.getId().equals(entity.getId())) { suffixes.add(e2.getMention()); } } suffixes.remove(t); suffixes.remove(entity.getTitle()); // String inputTextLower = inputText.toLowerCase(new // Locale("tr", // "TR")); // while (inputTextLower.contains(t)) { // int start = inputTextLower.indexOf(t); // int end = inputTextLower.indexOf(" ", start + t.length()); // if (end > start) { // String suffix = inputTextLower.substring(start, end); // // .replaceAll("\\W", ""); // if (suffix.contains("'") // || (Zemberek.getInstance().hasMorph(suffix) // && !suffix.equals(t) && suffix.length() > 4)) { // suffixes.add(suffix); // } // inputTextLower = inputTextLower.substring(end); // } else { // break; // } // } if (suffixes.size() >= minSuffix) { for (String suffix : suffixes) { double sim = gd.calculateSimilarity(suffix, type); suffixScore += sim; } } } // String entitySuffix = page.getSuffix(); // String[] inputSuffix = shingle.getSuffix().split(" "); // for (int j = 0; j < inputSuffix.length; j++) { // if (entitySuffix.contains(inputSuffix[j])) { // suffixScore += 0.25f; // } // } if (suffixScore > maxSuffixScore) { maxSuffixScore = suffixScore; } // if (id.equals("w691538")) { // LOGGER.info(""); // } double letterCaseScore = 0; int lc = page.getLetterCase(); if (StringUtils.isAllLowerCase(em.getMention()) && lc == 0 && id.startsWith("t")) { letterCaseScore = 1; } else if (StringUtils.isAllUpperCase(em.getMention()) && lc == 1 && id.startsWith("w")) { letterCaseScore = 1; } else if (Character.isUpperCase(em.getMention().charAt(0)) && lc == 2 && id.startsWith("w")) { letterCaseScore = 1; } else if (StringUtils.isAllLowerCase(em.getMention()) && id.startsWith("t")) { letterCaseScore = 1; } double nameScore = 1 - LevenshteinDistanceCalculator.calculateDistance(page.getTitle(), Zemberek.removeAfterSpostrophe(em.getMention())); double popularityScore = page.getRank(); if (id.startsWith("w")) { popularityScore = Math.log10(popularityScore + 1); if (popularityScore > maxPopularityScore) { maxPopularityScore = popularityScore; } } double leskScore = 0, simpleLeskScore = 0; String desc = em.getEntity().getPage().getDescription(); if (desc != null) { List<Token> tokens = Zemberek.getInstance().disambiguateFindTokens(desc, false, true); for (Token token : tokens) { if (inputTokensMultiset.contains(token.getMorphText()) && !TurkishNLP.isStopWord(token.getMorphText())) { simpleLeskScore += inputTokensMultiset.count(token.getMorphText()); } if (leskWords.contains(token.getMorphText()) && !TurkishNLP.isStopWord(token.getMorphText())) { leskScore += leskWords.count(token.getMorphText()); } } leskScore /= Math.log(tokens.size() + 1); simpleLeskScore /= Math.log(tokens.size() + 1); if (leskScore > maxLeskScore) { maxLeskScore = leskScore; } if (simpleLeskScore > maxSimpleLeskScore) { maxSimpleLeskScore = simpleLeskScore; } if (!entityScoreMap.containsKey(id)) { EntityScores scores = new EntityScores(em, id, popularityScore, nameScore, letterCaseScore, suffixScore, wordvecDescriptionScore, typeContentScore, typeScore, domainScore, hashDescriptionScore, wordvecDescriptionLocalScore, hashInfoboxScore, linkScore, wordvecLinksScore, leskScore, simpleLeskScore, typeClassifierScore); entityScoreMap.put(id, scores); } else { EntityScores entityScores = entityScoreMap.get(id); entityScores.setHashInfoboxScore((entityScores.getHashInfoboxScore() + hashInfoboxScore) / 2); entityScores.setHashDescriptionScore( (entityScores.getHashInfoboxScore() + hashDescriptionScore) / 2); entityScores.setLinkScore((entityScores.getLinkScore() + linkScore) / 2); entityScores.setWordvecDescriptionLocalScore( (entityScores.getWordvecDescriptionLocalScore() + wordvecDescriptionLocalScore) / 2); entityScores .setWordvecLinksScore((entityScores.getWordvecLinksScore() + wordvecLinksScore) / 2); entityScores.setLeskScore((entityScores.getLeskScore() + leskScore) / 2); } } } /* normalization and total score calculation starts */ Set<String> set = new HashSet<String>(); for (int i = 0; i < entities.size(); i++) { EntityMatch em = entities.get(i); String id = em.getId(); EntityScores entityScores = entityScoreMap.get(id); if (set.contains(id)) { continue; } if (id.startsWith("w")) { if (maxLinkScore > 0 && entityScores.getLinkScore() > 0) { entityScores.setLinkScore(entityScores.getLinkScore() / maxLinkScore); } if (maxHashInfoboxScore > 0 && entityScores.getHashInfoboxScore() > 0) { entityScores.setHashInfoboxScore(entityScores.getHashInfoboxScore() / maxHashInfoboxScore); } if (maxWordvecLinksScore > 0 && entityScores.getWordvecLinksScore() > 0) { entityScores.setWordvecLinksScore(entityScores.getWordvecLinksScore() / maxWordvecLinksScore); } if (maxPopularityScore > 0 && entityScores.getPopularityScore() > 0) { entityScores.setPopularityScore(entityScores.getPopularityScore() / maxPopularityScore); } } if (maxwordvecDescriptionLocalScore > 0 && entityScores.getWordvecDescriptionLocalScore() > 0) { entityScores.setWordvecDescriptionLocalScore( entityScores.getWordvecDescriptionLocalScore() / maxwordvecDescriptionLocalScore); } if (maxHashDescriptionScore > 0 && entityScores.getHashDescriptionScore() > 0) { entityScores .setHashDescriptionScore(entityScores.getHashDescriptionScore() / maxHashDescriptionScore); } if (maxWordvectorAverage > 0 && entityScores.getWordvecDescriptionScore() > 0) { entityScores.setWordvecDescriptionScore( entityScores.getWordvecDescriptionScore() / maxWordvectorAverage); } if (maxLeskScore > 0 && entityScores.getLeskScore() > 0) { entityScores.setLeskScore(entityScores.getLeskScore() / maxLeskScore); } if (maxSimpleLeskScore > 0 && entityScores.getSimpleLeskScore() > 0) { entityScores.setSimpleLeskScore(entityScores.getSimpleLeskScore() / maxSimpleLeskScore); } if (maxSuffixScore > 0 && entityScores.getSuffixScore() > 0) { entityScores.setSuffixScore(entityScores.getSuffixScore() / maxSuffixScore); } set.add(id); } LOGGER.info("\t" + "id\tTitle\tURL\tScore\tPopularity\tName\tLesk\tSimpeLesk\tCase\tNoun\tSuffix\tTypeContent\tType\tDomain\twordvecDescription\twordvecDescriptionLocal\thashDescription\thashInfobox\tword2vecLinks\tLink\t\ttypeClassifier\tDescription"); for (int i = 0; i < entities.size(); i++) { EntityMatch em = entities.get(i); String id = em.getId(); EntityScores e = entityScoreMap.get(id); double wikiScore = 0; if (id.startsWith("w") && Character.isUpperCase(em.getMention().charAt(0))) { wikiScore = wikiWeight; } else if (id.startsWith("t") && Character.isLowerCase(em.getMention().charAt(0))) { wikiScore = wikiWeight; } // if(id.equals("w508792")){ // LOGGER.info(""); // } double totalScore = wikiScore + e.getPopularityScore() * popularityWeight + e.getNameScore() * nameWeight + e.getLeskScore() * leskWeight + e.getSimpleLeskScore() * simpleLeskWeight + e.getLetterCaseScore() * letterCaseWeight + e.getSuffixScore() * suffixWeight + e.getTypeContentScore() * typeContentWeight + e.getTypeScore() * typeWeight + e.getDomainScore() * domainWeight + e.getWordvecDescriptionScore() * wordvecDescriptionWeight + e.getWordvecDescriptionLocalScore() * wordvecDescriptionLocalWeight + e.getHashDescriptionScore() * hashDescriptionWeight + e.getHashInfoboxScore() * hashInfoboxWeight + e.getWordvecLinksScore() * word2vecLinksWeight + e.getLinkScore() * linkWeight + e.getTypeClassifierkScore() * typeClassifierkWeight; if (ranklib == true) { totalScore = RankLib.getInstance().score(e); } if (em.getEntity().getPage().getUrlTitle().contains("(")) { totalScore /= 2; } em.setScore(totalScore); e.setScore(totalScore); LOGGER.info("\t" + id + "\t" + em.getEntity().getPage().getTitle() + "\t" + em.getEntity().getPage().getUrlTitle() + "\t" + em.getScore() + "\t" + e.getPopularityScore() * popularityWeight + "\t" + e.getNameScore() * nameWeight + "\t" + e.getLeskScore() * leskWeight + "\t" + e.getSimpleLeskScore() * simpleLeskWeight + "\t" + e.getLetterCaseScore() * letterCaseWeight + "\t" + e.getSuffixScore() * suffixWeight + "\t" + e.getTypeContentScore() * typeContentWeight + "\t" + e.getTypeScore() * typeWeight + "\t" + e.getDomainScore() * domainWeight + "\t" + e.getWordvecDescriptionScore() * wordvecDescriptionWeight + "\t" + e.getWordvecDescriptionLocalScore() * wordvecDescriptionLocalWeight + "\t" + e.getHashDescriptionScore() * hashDescriptionWeight + "\t" + e.getHashInfoboxScore() * hashInfoboxWeight + "\t" + e.getWordvecLinksScore() * word2vecLinksWeight + "\t" + e.getLinkScore() * linkWeight + "\t" + e.getTypeClassifierkScore() * typeClassifierkWeight + "\t" + em.getEntity().getPage().getDescription()); } // if (annotateEntities) { // annotateEntities(localParams.getParams().get("originalText"), sml); // } EntityMatchList eml = new EntityMatchList(); for (SpotMatch match : sml) { EntityMatchList list = match.getEntities(); if (!list.isEmpty()) { list.sort(); eml.add(list.get(0)); selectedEntities.add(list.get(0).getId()); } } return eml; }
From source file:org.aksw.simba.bengal.triple2nl.converter.SimpleIRIConverter.java
private String splitCamelCase(String s) { // we only split if it contains a vowel if (!(s.matches(".*[aeiou].*"))) { return s; }//from ww w .j a v a 2 s . c o m StringBuilder sb = new StringBuilder(); for (String token : s.split(" ")) { String[] tokenSplit = StringUtils.splitByCharacterTypeCamelCase(token); String noVowels = ""; for (String t : tokenSplit) { if (t.matches(".*[aeiou].*") || !StringUtils.isAllUpperCase(t)) { if (!noVowels.isEmpty()) { sb.append(noVowels).append(" "); noVowels = ""; } sb.append(t).append(" "); } else { noVowels += t; } // sb = new StringBuilder(sb.toString().trim()); } sb.append(noVowels); // sb.append(" "); } return sb.toString().trim(); // return s.replaceAll( // String.format("%s|%s|%s", // "(?<=[A-Z])(?=[A-Z][a-z])", // "(?<=[^A-Z])(?=[A-Z])", // "(?<=[A-Za-z])(?=[^A-Za-z])" // ), // " " // ); }