Lucene boosting only on matching field values

[echuta86]

Hi all,

I was wondering if there is any way to do a boost on the document based on
which value is in a field matched by a query.

ie, (Sample code below)
- You have a document that contains 1 field with multiple values.
- Field has value ABC boosted by 2.0
- Field has value XYZ boosted by 3.0
- I want the query that matches value XYZ to be ranked higher then the query
that matches value ABC

Is there a way to do this without slotting each value into a different
field?

Thanks for your time,
Eric

*Sample Code:*
import org.apache.lucene.document.Field;
import org.apache.lucene.analysis.standard.StandardAnalyzer;
import org.apache.lucene.queryParser.QueryParser;
import org.apache.lucene.search.Query;
import org.apache.lucene.store.Directory;
import org.apache.lucene.store.RAMDirectory;
import org.apache.lucene.document.Document;
import org.apache.lucene.index.IndexWriter;
import org.apache.lucene.search.TopDocs;
import org.apache.lucene.search.TopDocCollector;
import org.apache.lucene.search.Searcher;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.ScoreDoc;

public class LuceneScoreTest {

    public static void main (String[] args){

    try {

        // build index
        Directory index = buildIndex();
        Searcher searcher = new IndexSearcher(index);

        // query index
        String[] queries = new String[2];
        queries[0] = "+F1:(ABC)";
        queries[1] = "+F1:(XYZ)";

        for (int i=0 ; i<queries.length ; i++){
        String query = queries[i];
        System.out.println("==="+query+"===");
        Query q = new QueryParser("", new StandardAnalyzer()).parse(query);
        TopDocCollector hitCollector = new TopDocCollector(100);
        searcher.search(q,null,hitCollector);
        TopDocs topDocs = hitCollector.topDocs();

        // analyze results
        if (topDocs!=null && topDocs.totalHits>0){
            int totalHits = topDocs.totalHits;
            ScoreDoc[] docs = topDocs.scoreDocs;
            ScoreDoc doc;
            Document document;

System.out.println("topDocs.getMaxScore():"+topDocs.getMaxScore());
            System.out.println("topDocs.totalHits:"+totalHits);
            if (docs!=null && docs.length>0){
            for (int j = 0 ; j < docs.length ; j++){
                doc = docs[j];
                System.out.println("SCOREDOC[:"+j+"] score:"+doc.score);
                document = searcher.doc(doc.doc);
                System.out.println("DOCUMENT[:"+j+"]:"+document.get("F1"));
                System.out.println("DOCUMENT[:"+j+"]
boost:"+document.getBoost());
            }
            }
        }else{
            System.out.println("No docs found.");
        }
        }
    }catch (Exception e){
        e.printStackTrace();
    }
    }

    static Directory buildIndex () {
    Directory index = new RAMDirectory();
    IndexWriter w = null;
    Document doc = null;
    try {
        w = new IndexWriter(index, new StandardAnalyzer(), true);
        doc = new Document();
        Field f2 = new Field("F1", "XYZ", Field.Store.YES,
Field.Index.TOKENIZED);
        f2.setBoost(10.0f);
        doc.add(f2);
        Field f1 = new Field("F1", "ABC", Field.Store.YES,
Field.Index.TOKENIZED);
        f1.setBoost(1.0f);
        doc.add(f1);
        w.addDocument(doc);
        w.flush();
    }catch (Exception e){
        e.printStackTrace();
    }finally{
        try { if (w!=null) w.close(); }catch (Exception e){}
    }
    return index;
    }

}

[Grant Ingersoll-6]

Yes, see the Payload functionality and the BoostingTermQuery:  http://www.lucidimagination.com/search/?q=Payload

On Jul 9, 2009, at 6:42 PM, Eric Chu wrote:

> Hi all,
>
> I was wondering if there is any way to do a boost on the document  
> based on
> which value is in a field matched by a query.
>
> ie, (Sample code below)
> - You have a document that contains 1 field with multiple values.
> - Field has value ABC boosted by 2.0
> - Field has value XYZ boosted by 3.0
> - I want the query that matches value XYZ to be ranked higher then  
> the query
> that matches value ABC
>
> Is there a way to do this without slotting each value into a different
> field?
>
> Thanks for your time,
> Eric
>
> *Sample Code:*
> import org.apache.lucene.document.Field;
> import org.apache.lucene.analysis.standard.StandardAnalyzer;
> import org.apache.lucene.queryParser.QueryParser;
> import org.apache.lucene.search.Query;
> import org.apache.lucene.store.Directory;
> import org.apache.lucene.store.RAMDirectory;
> import org.apache.lucene.document.Document;
> import org.apache.lucene.index.IndexWriter;
> import org.apache.lucene.search.TopDocs;
> import org.apache.lucene.search.TopDocCollector;
> import org.apache.lucene.search.Searcher;
> import org.apache.lucene.search.IndexSearcher;
> import org.apache.lucene.search.ScoreDoc;
>
> public class LuceneScoreTest {
>
>    public static void main (String[] args){
>
>    try {
>
>        // build index
>        Directory index = buildIndex();
>        Searcher searcher = new IndexSearcher(index);
>
>        // query index
>        String[] queries = new String[2];
>        queries[0] = "+F1:(ABC)";
>        queries[1] = "+F1:(XYZ)";
>
>        for (int i=0 ; i<queries.length ; i++){
>        String query = queries[i];
>        System.out.println("==="+query+"===");
>        Query q = new QueryParser("", new  
> StandardAnalyzer()).parse(query);
>        TopDocCollector hitCollector = new TopDocCollector(100);
>        searcher.search(q,null,hitCollector);
>        TopDocs topDocs = hitCollector.topDocs();
>
>        // analyze results
>        if (topDocs!=null && topDocs.totalHits>0){
>            int totalHits = topDocs.totalHits;
>            ScoreDoc[] docs = topDocs.scoreDocs;
>            ScoreDoc doc;
>            Document document;
>
> System.out.println("topDocs.getMaxScore():"+topDocs.getMaxScore());
>            System.out.println("topDocs.totalHits:"+totalHits);
>            if (docs!=null && docs.length>0){
>            for (int j = 0 ; j < docs.length ; j++){
>                doc = docs[j];
>                System.out.println("SCOREDOC[:"+j+"]  
> score:"+doc.score);
>                document = searcher.doc(doc.doc);
>                System.out.println("DOCUMENT[:"+j
> +"]:"+document.get("F1"));
>                System.out.println("DOCUMENT[:"+j+"]
> boost:"+document.getBoost());
>            }
>            }
>        }else{
>            System.out.println("No docs found.");
>        }
>        }
>    }catch (Exception e){
>        e.printStackTrace();
>    }
>    }
>
>    static Directory buildIndex () {
>    Directory index = new RAMDirectory();
>    IndexWriter w = null;
>    Document doc = null;
>    try {
>        w = new IndexWriter(index, new StandardAnalyzer(), true);
>        doc = new Document();
>        Field f2 = new Field("F1", "XYZ", Field.Store.YES,
> Field.Index.TOKENIZED);
>        f2.setBoost(10.0f);
>        doc.add(f2);
>        Field f1 = new Field("F1", "ABC", Field.Store.YES,
> Field.Index.TOKENIZED);
>        f1.setBoost(1.0f);
>        doc.add(f1);
>        w.addDocument(doc);
>        w.flush();
>    }catch (Exception e){
>        e.printStackTrace();
>    }finally{
>        try { if (w!=null) w.close(); }catch (Exception e){}
>    }
>    return index;
>    }
>
> }

--------------------------
Grant Ingersoll
http://www.lucidimagination.com/

Search the Lucene ecosystem (Lucene/Solr/Nutch/Mahout/Tika/Droids)  
using Solr/Lucene:
http://www.lucidimagination.com/search


---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@...
For additional commands, e-mail: java-user-help@...