FileDocCategorySizeDatePackage
SynLookup.javaAPI DocApache Lucene 2.2.03595Sat Jun 16 22:20:46 BST 2007org.apache.lucene.wordnet

SynLookup

public class SynLookup extends Object
Test program to look up synonyms.

Fields Summary
Constructors Summary
Methods Summary
public static org.apache.lucene.search.Queryexpand(java.lang.String query, org.apache.lucene.search.Searcher syns, org.apache.lucene.analysis.Analyzer a, java.lang.String field, float boost)
Perform synonym expansion on a query.

param
query
param
syns
param
a
param
field
param
boost

		Set already = new HashSet(); // avoid dups		
		List top = new LinkedList(); // needs to be separately listed..

		// [1] Parse query into separate words so that when we expand we can avoid dups
		TokenStream ts = a.tokenStream( field, new StringReader( query));
		org.apache.lucene.analysis.Token t;
		while ( (t = ts.next()) != null)
		{
			String word = t.termText();
			if ( already.add( word))
				top.add( word);
		}
		BooleanQuery tmp = new BooleanQuery();
		
		// [2] form query
		Iterator it = top.iterator();
		while ( it.hasNext())
		{
			// [2a] add to level words in
			String word = (String) it.next();
			TermQuery tq = new TermQuery( new Term( field, word));
			tmp.add( tq, BooleanClause.Occur.SHOULD);

			// [2b] add in unique synonums
			Hits hits = syns.search( new TermQuery( new Term(Syns2Index.F_WORD, word)));
			for (int i = 0; i < hits.length(); i++)
			{
				Document doc = hits.doc(i);
				String[] values = doc.getValues( Syns2Index.F_SYN);
				for ( int j = 0; j < values.length; j++)
				{
					String syn = values[ j];
					if ( already.add( syn))
					{
						tq = new TermQuery( new Term( field, syn));
						if ( boost > 0) // else keep normal 1.0
							tq.setBoost( boost);
						tmp.add( tq, BooleanClause.Occur.SHOULD); 
					}
				}
			}
		}


		return tmp;
	
public static voidmain(java.lang.String[] args)

		if (args.length != 2) {
			System.out.println(
							   "java org.apache.lucene.wordnet.SynLookup <index path> <word>");
		}

		FSDirectory directory = FSDirectory.getDirectory(args[0], false);
		IndexSearcher searcher = new IndexSearcher(directory);

		String word = args[1];
		Hits hits = searcher.search(
									new TermQuery(new Term(Syns2Index.F_WORD, word)));

		if (hits.length() == 0) {
			System.out.println("No synonyms found for " + word);
		} else {
			System.out.println("Synonyms found for \"" + word + "\":");
		}

		for (int i = 0; i < hits.length(); i++) {
			Document doc = hits.doc(i);

			String[] values = doc.getValues(Syns2Index.F_SYN);

			for (int j = 0; j < values.length; j++) {
				System.out.println(values[j]);
			}
		}

		searcher.close();
		directory.close();