Java MoreLikeThis.setFieldNames方法代码示例

import org.apache.lucene.queries.mlt.MoreLikeThis; //导入方法依赖的package包/类
public synchronized Sample getSample(String id) throws Exception {
	Term term = new Term(Sample.ID, id);
	TermQuery tq = new TermQuery(term);
	TopDocs td = getIndexSearcher().search(tq, 1);
	if (td.totalHits > 0) {
		Document doc = indexSearcher.doc(td.scoreDocs[0].doc);
		Sample s = (Sample) SerializationUtil.deserialize(doc.getBinaryValue("RawData").bytes);

		MoreLikeThis mlt = new MoreLikeThis(indexSearcher.getIndexReader());
		mlt.setFieldNames(new String[] { Variable.LABEL, Variable.DESCRIPTION, Variable.TAGS });
		mlt.setMaxWordLen(MAXWORDLENGTH);
		String[] terms = mlt.retrieveInterestingTerms(td.scoreDocs[0].doc);
		for (int i = 0; i < 10 && i < terms.length; i++) {
			s.put(Variable.SUGGESTEDTAGS, terms[i]);
		}

		return s;
	} else {
		return null;
	}
}
 

import org.apache.lucene.queries.mlt.MoreLikeThis; //导入方法依赖的package包/类
public synchronized ListDataSet searchSimilar(Sample sample, int start, int count) throws Exception {
	Term term = new Term(Sample.ID, sample.getId());
	TermQuery tq = new TermQuery(term);
	TopDocs td = getIndexSearcher().search(tq, count);
	if (td == null || td.totalHits == 0) {
		ListDataSet ds = ListDataSet.Factory.emptyDataSet();
		return ds;
	}
	MoreLikeThis mlt = new MoreLikeThis(indexSearcher.getIndexReader());
	mlt.setFieldNames(new String[] { Variable.LABEL, Variable.DESCRIPTION, Variable.TAGS });
	mlt.setMaxWordLen(MAXWORDLENGTH);
	Query query = mlt.like(td.scoreDocs[0].doc);
	BooleanQuery bq = new BooleanQuery();
	bq.add(query, Occur.MUST);
	bq.add(new TermQuery(new Term("Id", sample.getId())), Occur.MUST_NOT);

	return search(bq, start, count);
}
 

import org.apache.lucene.queries.mlt.MoreLikeThis; //导入方法依赖的package包/类
/**
 * {@inheritDoc}
 */
@Override
public void train(AtomicReader atomicReader, String[] textFieldNames, String classFieldName, Analyzer analyzer, Query query) throws IOException {
  this.textFieldNames = textFieldNames;
  this.classFieldName = classFieldName;
  mlt = new MoreLikeThis(atomicReader);
  mlt.setAnalyzer(analyzer);
  mlt.setFieldNames(textFieldNames);
  indexSearcher = new IndexSearcher(atomicReader);
  if (minDocsFreq > 0) {
    mlt.setMinDocFreq(minDocsFreq);
  }
  if (minTermFreq > 0) {
    mlt.setMinTermFreq(minTermFreq);
  }
  this.query = query;
}
 

import org.apache.lucene.queries.mlt.MoreLikeThis; //导入方法依赖的package包/类
/**
 * {@inheritDoc}
 */
@Override
public void train(AtomicReader atomicReader, String textFieldName, String classFieldName, Analyzer analyzer) throws IOException {
  this.textFieldName = textFieldName;
  this.classFieldName = classFieldName;
  mlt = new MoreLikeThis(atomicReader);
  mlt.setAnalyzer(analyzer);
  mlt.setFieldNames(new String[]{textFieldName});
  indexSearcher = new IndexSearcher(atomicReader);
}
 

import org.apache.lucene.queries.mlt.MoreLikeThis; //导入方法依赖的package包/类
/**
 * {@inheritDoc}
 */
@Override
public void train(AtomicReader atomicReader, String textFieldName, String classFieldName, Analyzer analyzer, Query query) throws IOException {
  this.textFieldNames = new String[]{textFieldName};
  this.classFieldName = classFieldName;
  mlt = new MoreLikeThis(atomicReader);
  mlt.setAnalyzer(analyzer);
  mlt.setFieldNames(new String[]{textFieldName});
  indexSearcher = new IndexSearcher(atomicReader);
  this.query = query;
}
 

import org.apache.lucene.queries.mlt.MoreLikeThis; //导入方法依赖的package包/类
public static void main(String[] args) throws Throwable {

        String indexDir = System.getProperty("index.dir");
        FSDirectory directory = FSDirectory.open(new File(indexDir));
        IndexReader reader = DirectoryReader.open(directory);

        IndexSearcher searcher = new IndexSearcher(reader);

        int numDocs = reader.maxDoc();

        MoreLikeThis mlt = new MoreLikeThis(reader);
        mlt.setFieldNames(new String[] { "title", "author" });
        mlt.setMinTermFreq(1);
        mlt.setMinDocFreq(1);

        for (int docID = 0; docID < numDocs; docID++) {
            LOGGER.info();
            StoredDocument doc = reader.document(docID);
            LOGGER.info(doc.get("title"));

            Query query = mlt.like(docID);
            LOGGER.info("  query=" + query);

            TopDocs similarDocs = searcher.search(query, 10);
            if (similarDocs.totalHits == 0)
                LOGGER.info("  None like this");
            for (int i = 0; i < similarDocs.scoreDocs.length; i++) {
                if (similarDocs.scoreDocs[i].doc != docID) {
                    doc = reader.document(similarDocs.scoreDocs[i].doc);
                    LOGGER.info("  -> " + doc.getField("title").stringValue());
                }
            }
        }

        reader.close();
        directory.close();
    }
 

import org.apache.lucene.queries.mlt.MoreLikeThis; //导入方法依赖的package包/类
/**
 * Searches for similar {@link News}.
 * @return a {@link List} with the similar {@link News}. Can be empty.
 * 		The more similar the more top is the {@link News} in the {@link List}.
 */
public List<News> getSimilar() {
	final int docId = IndexSearch.getInstance().getDocIdForId(
			NewsIndexType.getInstance(), String.valueOf(news.getId()));

	// configure "more like this"
	final MoreLikeThis moreLikeThis = IndexSearch.getInstance().newMoreLikeThis(news.getLocale());
	moreLikeThis.setMinWordLen(3);
	moreLikeThis.setBoost(true);
	moreLikeThis.setBoostFactor(10);
	moreLikeThis.setFieldNames(new String[] {
			NewsIndexType.FIELD_TITLE,
			NewsIndexType.FIELD_DESCRIPTION
	});

	try {
		final BooleanQuery query = new BooleanQuery();

		// it must have the same locale
		QueryUtil.addLocale(query, news.getLocale());

		// filter with publish date
		final NumericRangeQuery<Long> dateQuery = NumericRangeQuery.newLongRange(
				NewsIndexType.FIELD_PUBLISH_DATE, getDate(-PUBLISH_DATE_DELTA),
				getDate(PUBLISH_DATE_DELTA), true, true);
		query.add(dateQuery, Occur.MUST);

		// it must be in the same category.
		final NumericRangeQuery<Integer> categoryQuery = NumericRangeQuery.newIntRange(
				NewsIndexType.FIELD_CATEGORYID, news.getCategoryId(), news.getCategoryId(), true, true);
		query.add(categoryQuery, Occur.MUST);

		final Query moreQuery = moreLikeThis.like(docId);
		query.add(moreQuery, Occur.MUST);

		// not the same news
		query.add(new TermQuery(new Term(IIndexElement.FIELD_ID, String.valueOf(news.getId()))), Occur.MUST_NOT);

		// execute query
		final DocumentsSearchResult result = IndexSearch.getInstance().search(query, new SearchOptions());

		final List<Document> resultDocs = new ArrayList<>();
		for (final Document doc : result.getResults()) {
			final float score = result.getScore(doc);

			// use only news with a sufficient score.
			if(score >= MIN_SCORE) {
				resultDocs.add(doc);
				LOGGER.debug("News {} is similar to news {}.", doc.get(IIndexElement.FIELD_ID), news.getId());
			}
			else {
				LOGGER.debug("News {} has not a sufficient score to be similar to news {}.",
						doc.get(IIndexElement.FIELD_ID), news.getId());
			}
		}

		// convert the Documents to News.
		return NewsIndexType.docsToNews(resultDocs);
	}
	catch (final IOException e) {
		LOGGER.error("Can't build query for news with id '{}'.", news.getId());
	}

	return null;
}