lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Koji Sekiguchi <k...@r.email.ne.jp>
Subject Re: Regarding MoreLikeThis similarity Search
Date Sat, 19 Mar 2011 00:49:15 GMT
(11/03/19 6:16), madhuri_1820@yahoo.com wrote:
> Hi,
>
> I am new to lucene ... I have a question while implementing similarity search using MoreLikeThis
query. I have written a small program but it is not giving any results. In my index file I
have both strored and unstored(analyzed) fields.
>
> Sample Code :
>
> IndexReader ir =  IndexReader.open(index directory path);
> 	    		IndexSearcher indexsearcher = new ProfileIndexSearcher(ir);
> 	    		IndexSearcher indexsearcher = new IndexSearcher(ir);
> 		    	MoreLikeThis mlt = new MoreLikeThis(ir);
> 		    	Analyzer analyzer = new StandardAnalyzer(Version.LUCENE_30);
> 	    		mlt.setAnalyzer(analyzer);
> 	    		mlt.setFieldNames(string_of_fields);
> 	    		
> 		    	Query query1 = mlt.like(1); // mlt.like(docnumber)
> 		    	TopDocs matches = null;
> 		    	matches = indexsearcher.search(query1,20);
>
> As per my understanding I should atleast one document as result.
>
> Is any one has idea what I can do to implement this serach.
> Thanks in Advance.
>
> --Madhu.

Hi Madhu,

Check query1 and compare what you expected:

Query query1 = mlt.like(1); // mlt.like(docnumber)
System.out.println(query1.toString());

And if it is not what you expected, check min/max docFreq/termFreq/termLength
setter/getter methods described in MLT javadoc.

Koji
-- 
http://www.rondhuit.com/en/

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


Mime
View raw message