Examples with TextAnnotation edu.illinois.cs.cogcomp.core.datastructures.textannotation.TextAnnotation used on opensource projects

Example 1 with TextAnnotation

use of edu.illinois.cs.cogcomp.core.datastructures.textannotation.TextAnnotation in project cogcomp-nlp by CogComp.

the class ServerClientAnnotator method annotate.

public TextAnnotation annotate(String str) throws Exception {
    String viewsConnected = Arrays.toString(viewsToAdd);
    String views = viewsConnected.substring(1, viewsConnected.length() - 1).replace(" ", "");
    ConcurrentMap<String, byte[]> concurrentMap = (db != null) ? db.hashMap(viewName, Serializer.STRING, Serializer.BYTE_ARRAY).createOrOpen() : null;
    String key = DigestUtils.sha1Hex(str + views);
    if (concurrentMap != null && concurrentMap.containsKey(key)) {
        byte[] taByte = concurrentMap.get(key);
        return SerializationHelper.deserializeTextAnnotationFromBytes(taByte);
    } else {
        URL obj = new URL(url + ":" + port + "/annotate");
        HttpURLConnection con = (HttpURLConnection) obj.openConnection();
        con.setRequestMethod("POST");
        con.setRequestProperty("charset", "utf-8");
        con.setRequestProperty("Content-Type", "text/plain; charset=utf-8");
        con.setDoOutput(true);
        con.setUseCaches(false);
        OutputStreamWriter wr = new OutputStreamWriter(con.getOutputStream());
        wr.write("text=" + URLEncoder.encode(str, "UTF-8") + "&views=" + views);
        wr.flush();
        int responseCode = con.getResponseCode();
        logger.debug("\nSending '" + con.getRequestMethod() + "' request to URL : " + url);
        logger.debug("Response Code : " + responseCode);
        BufferedReader in = new BufferedReader(new InputStreamReader(con.getInputStream()));
        String inputLine;
        StringBuilder response = new StringBuilder();
        while ((inputLine = in.readLine()) != null) {
            response.append(inputLine);
        }
        in.close();
        TextAnnotation ta = SerializationHelper.deserializeFromJson(response.toString());
        if (concurrentMap != null) {
            concurrentMap.put(key, SerializationHelper.serializeTextAnnotationToBytes(ta));
            this.db.commit();
        }
        return ta;
    }
}

Example 2 with TextAnnotation

use of edu.illinois.cs.cogcomp.core.datastructures.textannotation.TextAnnotation in project cogcomp-nlp by CogComp.

the class CurrencyIndicator method getFeatures.

@Override
public Set<Feature> getFeatures(Constituent c) throws EdisonException {
    try {
        if (!loaded)
            synchronized (this) {
                // now its changed to be loaded from datastore.
                if (!loaded)
                    loadCurrency(gzip, true);
            }
    } catch (Exception ex) {
        throw new EdisonException(ex);
    }
    TextAnnotation ta = c.getTextAnnotation();
    if (!ta.hasView(VIEW_NAME)) {
        try {
            addCurrencyView(ta);
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    SpanLabelView view = (SpanLabelView) ta.getView(VIEW_NAME);
    Set<Feature> features = new LinkedHashSet<>();
    for (Constituent cc : view.where(Queries.containedInConstituent(c))) {
        if (cc.getEndSpan() == c.getEndSpan()) {
            if (cc.getStartSpan() - 1 > c.getEndSpan()) {
                // check if this is a number
                if (WordLists.NUMBERS.contains(ta.getToken(cc.getStartSpan() - 1).toLowerCase())) {
                    features.add(CURRENCY);
                    break;
                }
            }
        } else if (WordFeatureExtractorFactory.numberNormalizer.getWordFeatures(ta, cc.getEndSpan()).size() > 0) {
            features.add(CURRENCY);
            break;
        }
    }
    return features;
}

Example 3 with TextAnnotation

use of edu.illinois.cs.cogcomp.core.datastructures.textannotation.TextAnnotation in project cogcomp-nlp by CogComp.

the class ParseHeadWordFeatureExtractor method getFeatures.

@Override
public Set<Feature> getFeatures(Constituent c) throws EdisonException {
    TextAnnotation ta = c.getTextAnnotation();
    TreeView tree = (TreeView) ta.getView(parseViewName);
    Constituent phrase;
    try {
        phrase = tree.getParsePhrase(c);
    } catch (Exception e) {
        throw new EdisonException(e);
    }
    Set<Feature> features = new LinkedHashSet<>();
    int head = CollinsHeadFinder.getInstance().getHeadWordPosition(phrase);
    Constituent c1 = new Constituent("", "", ta, head, head + 1);
    features.addAll(fex.getFeatures(c1));
    return features;
}

Example 4 with TextAnnotation

use of edu.illinois.cs.cogcomp.core.datastructures.textannotation.TextAnnotation in project cogcomp-nlp by CogComp.

the class ContextFeatureExtractor method getFeatures.

@Override
public Set<Feature> getFeatures(Constituent c) throws EdisonException {
    TextAnnotation ta = c.getTextAnnotation();
    int start = c.getStartSpan() - contextSize;
    int end = c.getEndSpan() + contextSize;
    if (start < 0)
        start = 0;
    if (end >= ta.size())
        end = ta.size();
    Set<Feature> features = new LinkedHashSet<>();
    for (int i = start; i < end; i++) {
        if (ignoreConstituent)
            if (c.getStartSpan() <= i && i < c.getEndSpan())
                continue;
        for (FeatureExtractor f : this.generators) {
            Constituent neighbor = new Constituent("TMP", "TMP", ta, i, i + 1);
            Set<Feature> feats = f.getFeatures(neighbor);
            for (Feature feat : feats) {
                String preamble = "context";
                if (specifyIndex) {
                    String index = "*";
                    if (i < c.getStartSpan())
                        index = (i - c.getStartSpan()) + "";
                    else if (i >= c.getEndSpan())
                        index = (i - c.getEndSpan() + 1) + "";
                    preamble += index;
                }
                preamble += ":";
                features.add(feat.prefixWith(preamble + f.getName()));
            }
        }
    }
    return features;
}

Example 5 with TextAnnotation

use of edu.illinois.cs.cogcomp.core.datastructures.textannotation.TextAnnotation in project cogcomp-nlp by CogComp.

the class BIOTester method statistics.

public static void statistics() {
    int ace_nam = 0;
    int ace_nom = 0;
    int ace_pro = 0;
    int ere_nam = 0;
    int ere_nom = 0;
    int ere_pro = 0;
    int tac_nam = 0;
    int tac_nom = 0;
    try {
        ACEReaderWithTrueCaseFixer aceReader = new ACEReaderWithTrueCaseFixer("data/all", false);
        for (TextAnnotation ta : aceReader) {
            for (Constituent c : ta.getView(ViewNames.MENTION_ACE)) {
                if (c.getAttribute("EntityMentionType").equals("NAM")) {
                    ace_nam++;
                }
                if (c.getAttribute("EntityMentionType").equals("NOM")) {
                    ace_nom++;
                }
                if (c.getAttribute("EntityMentionType").equals("PRO")) {
                    ace_pro++;
                }
            }
        }
        EREMentionRelationReader ereReader = new EREMentionRelationReader(EREDocumentReader.EreCorpus.ENR3, "data/ere/data", false);
        for (XmlTextAnnotation xta : ereReader) {
            TextAnnotation ta = xta.getTextAnnotation();
            for (Constituent c : ta.getView(ViewNames.MENTION_ERE)) {
                if (c.getAttribute("EntityMentionType").equals("NAM")) {
                    ere_nam++;
                }
                if (c.getAttribute("EntityMentionType").equals("NOM")) {
                    ere_nom++;
                }
                if (c.getAttribute("EntityMentionType").equals("PRO")) {
                    ere_pro++;
                }
            }
        }
        ColumnFormatReader columnFormatReader = new ColumnFormatReader("data/tac/2016.nam");
        for (TextAnnotation ta : columnFormatReader) {
            for (Constituent c : ta.getView("MENTIONS")) {
                tac_nam++;
            }
        }
        columnFormatReader = new ColumnFormatReader("data/tac/2016.nom");
        for (TextAnnotation ta : columnFormatReader) {
            for (Constituent c : ta.getView("MENTIONS")) {
                tac_nom++;
            }
        }
    } catch (Exception e) {
        e.printStackTrace();
    }
    System.out.println("ACE_NAM: " + ace_nam);
    System.out.println("ACE_NOM: " + ace_nom);
    System.out.println("ACE_PRO: " + ace_pro);
    System.out.println("ERE_NAM: " + ere_nam);
    System.out.println("ERE_NOM: " + ere_nom);
    System.out.println("ERE_PRO: " + ere_pro);
    System.out.println("TAC_NAM: " + tac_nam);
    System.out.println("TAC_NOM: " + tac_nom);
}