Example usage for org.apache.lucene.document Document getValues

Introduction

In this page you can find the example usage for org.apache.lucene.document Document getValues.

Prototype

public final String[] getValues(String name) 

Source Link

Document

Returns an array of values of the field specified as the method parameter.

Usage

From source file:TestWang.java

License:Open Source License

public void computeMAP(ImageSearcher searcher, String prefix) throws IOException {
    System.out.println("ComputeMap");
    // copy index to ram to be much faster ...
    IndexReader reader = IndexReader.open(new RAMDirectory(FSDirectory.open(new File(indexPath))), true);

    Pattern p = Pattern.compile("([0-9]+).jpg");
    double map = 0;
    double errorRate = 0d;
    double precision10 = 0d;
    double[] pr10cat = new double[10];
    double[] pr10cnt = new double[10];
    for (int i = 0; i < pr10cat.length; i++) {
        pr10cat[i] = 0d;/*from w ww .j  av a 2 s  . c  om*/
        pr10cnt[i] = 0d;
    }
    System.out.println("name\tmap\tp@10\terror rate");
    System.out.println("Anzahl Anfragen: " + sampleQueries.length);
    for (int i = 0; i < sampleQueries.length; i++) {
        int id = sampleQueries[i];
        String file = testExtensive + "/" + id + ".jpg";
        ImageSearchHits hits = searcher.search(findDoc(reader, id + ".jpg"), reader);
        int goodOnes = 0;
        double avgPrecision = 0d;
        double precision10temp = 0d;
        int countResults = 0;
        //            System.out.println("Anzahl Hits: " + hits.length());
        for (int j = 0; j < hits.length(); j++) {
            Document d = hits.doc(j);
            String hitsId = d.getValues(DocumentBuilder.FIELD_NAME_IDENTIFIER)[0];
            Matcher matcher = p.matcher(hitsId);
            if (matcher.find())
                hitsId = matcher.group(1);
            else
                fail("Did not get the number ...");
            int testID = Integer.parseInt(hitsId);
            if (testID != id)
                countResults++;
            if ((testID != id) && ((int) Math.floor(id / 100) == (int) Math.floor(testID / 100))) {
                goodOnes++;
                // Only if there is a change in recall
                avgPrecision += (double) goodOnes / (double) countResults;
                //                    System.out.print("x");
                if (j <= 10) {
                    precision10temp += 1d;
                }
            } else {
                if (j == 1) { // error rate
                    errorRate++;
                }
            }
        } // end for loop iterating results.
        //            if (avgPrecision<=0) {
        //                System.out.println("avgPrecision = " + avgPrecision);
        //                System.out.println("goodOnes = " + goodOnes);
        //            }
        //            assertTrue("Check if average precision is > 0", avgPrecision > 0);
        //            assertTrue("Check if goodOnes is > 0", goodOnes > 0);
        avgPrecision = avgPrecision / goodOnes;
        precision10 += precision10temp / 10d;
        // precision @ 10 for each category ...
        pr10cat[(int) Math.floor(id / 100)] += precision10temp / 10d;
        pr10cnt[(int) Math.floor(id / 100)] += 1d;
        map += avgPrecision;
    }
    map = map / sampleQueries.length;
    errorRate = errorRate / sampleQueries.length;
    precision10 = precision10 / sampleQueries.length;
    System.out.print(prefix + "\t");
    System.out.print(String.format("%.5f ", map) + '\t');
    System.out.print(String.format("%.5f ", precision10) + '\t');
    System.out.print(String.format("%.5f ", errorRate) + '\t');
    // precision@10 per category
    for (int i = 0; i < pr10cat.length; i++) {
        double v = 0;
        if (pr10cnt[i] > 0)
            v = pr10cat[i] / pr10cnt[i];
        //            System.out.print(i + ": ");
        System.out.printf("%.5f\t", v);

    }
    System.out.println();
}

---

From source file:TestWang.java

License:Open Source License

private Document findDoc(IndexReader reader, String file) throws IOException {
    for (int i = 0; i < reader.numDocs(); i++) {
        Document document = reader.document(i);
        String s = document.getValues(DocumentBuilder.FIELD_NAME_IDENTIFIER)[0];
        if (s.endsWith(File.separator + file)) {
            //                System.out.println("s = " + s);
            return document;
        }/* ww w  .ja  va 2s  .c o  m*/
    }
    return null;
}

---

From source file:TestWang.java

License:Open Source License

private Document[] findDocs(IndexReader reader, String[] file) throws IOException {
    Document[] result = new Document[file.length];
    for (int i = 0; i < reader.numDocs(); i++) {
        Document document = reader.document(i);
        String s = document.getValues(DocumentBuilder.FIELD_NAME_IDENTIFIER)[0];
        for (int j = 0; j < result.length; j++) {
            if (s.endsWith("\\" + file[j])) {
                //                System.out.println("s = " + s);
                result[j] = document;/*from  w ww.j  a va2s. co m*/
            }
        }
    }
    return result;
}

---

From source file:KNearestNeighbourDocumentClassifier.java

License:Apache License

/**
 * Returns the top k results from a More Like This query based on the input document
 *
 * @param document the document to use for More Like This search
 * @return the top results for the MLT query
 * @throws IOException If there is a low-level I/O error
 *///from  ww  w.  ja  v  a  2  s .c o  m
private TopDocs knnSearch(Document document) throws IOException {
    BooleanQuery.Builder mltQuery = new BooleanQuery.Builder();

    for (String fieldName : textFieldNames) {
        String boost = null;
        if (fieldName.contains("^")) {
            String[] field2boost = fieldName.split("\\^");
            fieldName = field2boost[0];
            boost = field2boost[1];
        }
        String[] fieldValues = document.getValues(fieldName);
        if (boost != null) {
            mlt.setBoost(true);
            mlt.setBoostFactor(Float.parseFloat(boost));
        }
        mlt.setAnalyzer(field2analyzer.get(fieldName));
        for (String fieldContent : fieldValues) {
            mltQuery.add(new BooleanClause(mlt.like(fieldName, new StringReader(fieldContent)),
                    BooleanClause.Occur.SHOULD));
        }
        mlt.setBoost(false);
    }
    Query classFieldQuery = new WildcardQuery(new Term(classFieldName, "*"));
    mltQuery.add(new BooleanClause(classFieldQuery, BooleanClause.Occur.MUST));
    if (query != null) {
        mltQuery.add(query, BooleanClause.Occur.MUST);
    }
    return indexSearcher.search(mltQuery.build(), k);
}

---

From source file:aos.lucene.analysis.synonym.WordNetSynonymEngine.java

License:Apache License

public String[] getSynonyms(String word) throws IOException {

    List<String> synList = new ArrayList<String>();

    AllDocCollector collector = new AllDocCollector(); // #A

    searcher.search(new TermQuery(new Term("word", word)), collector);

    for (ScoreDoc hit : collector.getHits()) { // #B
        Document doc = searcher.doc(hit.doc);

        String[] values = doc.getValues("syn");

        for (String syn : values) { // #C
            synList.add(syn);/*w w w . j  a  v a 2 s .c  o  m*/
        }
    }

    return synList.toArray(new String[0]);
}

---

From source file:aos.lucene.search.advanced.BooksLikeThis.java

License:Apache License

public Document[] docsLike(int id, int max) throws IOException {
    Document doc = reader.document(id);

    String[] authors = doc.getValues("author");
    BooleanQuery authorQuery = new BooleanQuery(); //
    for (String author : authors) { //
        authorQuery.add(new TermQuery(new Term("author", author)), //
                BooleanClause.Occur.SHOULD); //
    }/*from   w ww .j a v  a 2 s.  c o  m*/
    authorQuery.setBoost(2.0f);

    TermFreqVector vector = //
            reader.getTermFreqVector(id, "subject"); //

    BooleanQuery subjectQuery = new BooleanQuery(); //
    for (String vecTerm : vector.getTerms()) { //
        TermQuery tq = new TermQuery( //
                new Term("subject", vecTerm)); //
        subjectQuery.add(tq, BooleanClause.Occur.SHOULD); //
    }

    BooleanQuery likeThisQuery = new BooleanQuery(); //
    likeThisQuery.add(authorQuery, BooleanClause.Occur.SHOULD); //
    likeThisQuery.add(subjectQuery, BooleanClause.Occur.SHOULD); //

    likeThisQuery.add(new TermQuery( //
            new Term("isbn", doc.get("isbn"))), BooleanClause.Occur.MUST_NOT); //

    // LOGGER.info("  Query: " +
    // likeThisQuery.toString("contents"));
    TopDocs hits = searcher.search(likeThisQuery, 10);
    int size = max;
    if (max > hits.scoreDocs.length)
        size = hits.scoreDocs.length;

    Document[] docs = new Document[size];
    for (int i = 0; i < size; i++) {
        docs[i] = reader.document(hits.scoreDocs[i].doc);
    }

    return docs;
}

---

From source file:at.ac.univie.mminf.luceneSKOS.analysis.engine.jena.SKOSEngineImpl.java

License:Apache License

@Override
public Collection<String> getConcepts(String label) throws IOException {
    Set<String> concepts = new HashSet<>();
    // convert the query to lower-case
    String queryString = label.toLowerCase(Locale.ROOT);
    AllDocCollector collector = new AllDocCollector();
    DisjunctionMaxQuery query = new DisjunctionMaxQuery(0.0f);
    query.add(new TermQuery(new Term(FIELD_PREF_LABEL, queryString)));
    query.add(new TermQuery(new Term(FIELD_ALT_LABEL, queryString)));
    query.add(new TermQuery(new Term(FIELD_HIDDEN_LABEL, queryString)));
    searcher.search(query, collector);/*from  ww w .  j a  va2 s . c  o m*/
    for (Integer hit : collector.getDocs()) {
        Document doc = searcher.doc(hit);
        String conceptURI = doc.getValues(FIELD_URI)[0];
        concepts.add(conceptURI);
    }
    return concepts;
}

---

From source file:at.ac.univie.mminf.luceneSKOS.analysis.engine.jena.SKOSEngineImpl.java

License:Apache License

/**
 * Returns the values of a given field for a given concept
 */// ww w .  jav  a2  s  .  com
private Collection<String> readConceptFieldValues(String conceptURI, String field) throws IOException {
    Query query = new TermQuery(new Term(FIELD_URI, conceptURI));
    TopDocs docs = searcher.search(query, 1);
    ScoreDoc[] results = docs.scoreDocs;
    if (results.length != 1) {
        logger.warn("Unknown concept " + conceptURI);
        return null;
    }
    Document conceptDoc = searcher.doc(results[0].doc);
    return Arrays.asList(conceptDoc.getValues(field));
}

---

From source file:at.ac.univie.mminf.luceneSKOS.analysis.SKOSURIFilterTest.java

License:Apache License

@Test
public void singleUriExpansionWithStoredField() throws CorruptIndexException, IOException {

    Document doc = new Document();
    doc.add(new Field("subject", "http://example.com/concept/1", Field.Store.YES, Field.Index.ANALYZED));

    writer.addDocument(doc);//from  w w w  .j  a va 2 s .  co m

    searcher = new IndexSearcher(IndexReader.open(writer, false));

    Query query = new TermQuery(new Term("subject", "leaps"));

    TopDocs results = searcher.search(query, 10);
    Assert.assertEquals(1, results.totalHits);

    Document indexDoc = searcher.doc(results.scoreDocs[0].doc);

    String[] fieldValues = indexDoc.getValues("subject");

    Assert.assertEquals(1, fieldValues.length);

    Assert.assertEquals(fieldValues[0], "http://example.com/concept/1");

}

---

From source file:at.ac.univie.mminf.luceneSKOS.analysis.SKOSURIFilterTest.java

License:Apache License

@Test
public void singleUriExpansionWithUnstoredField() throws CorruptIndexException, IOException {

    Document doc = new Document();
    doc.add(new Field("subject", "http://example.com/concept/1", Field.Store.NO, Field.Index.ANALYZED));

    writer.addDocument(doc);//  ww  w  . ja va  2  s.c  o m

    searcher = new IndexSearcher(IndexReader.open(writer, false));

    Query query = new TermQuery(new Term("subject", "jumps"));

    TopDocs results = searcher.search(query, 10);
    Assert.assertEquals(1, results.totalHits);

    Document indexDoc = searcher.doc(results.scoreDocs[0].doc);

    String[] fieldValues = indexDoc.getValues("subject");

    Assert.assertEquals(0, fieldValues.length);

}

---

• «
• 1
• 2
• 3
• 4
• 5
• 6
• 7
• 8
• 9
• 10
• 11
• 12
• 13
• 14
• 15
• 16
• 17
• 18
• 19
• 20
• 21
• 22
• 23
• »