Document Score...

[Peter]

How do you get the document score - I believe this is how well it matches a query? - when getting the search results as a Java list?

Je l'essayerai en franais...
Comment obtenez-vous les points de document - est-ce que je crois que c'est  quel point lui des allumettes une question?  - quand obtenir la recherche rsulte-t-il comme Java List?

[Rida Benjelloun]

Hi,
Actually, you can't get the score of the hit in a java list. I will try to fix the problem.
You can use this methode :
public static void search(File indexDir, String queryStr)  throws Exception{
  Directory fsDir = FSDirectory.getDirectory(indexDir, false);
  IndexSearcher is = new IndexSearcher(fsDir);

  Query query = QueryParser.parse(queryStr, "contents", new StandardAnalyzer());
  Hits hits = is.search(query);
  System.out.println("Found " + hits.length() +
                    " document(s) that matched query '" + q + "':");
  for (int i = 0; i < hits.length(); i++) {
    Document doc = hits.doc(i);

    System.out.println("score : "+ hits.score(i));
    System.out.println(doc.get("filename"));
  }
}

[Peter]

Thank you for your response. 

I've been looking through the lucene documentation and I've had a lot of trouble finding documentation to explain the document score.  All I've found is that the score is a number of type float between 1 and 0, which is somehow normalized based on the length of the content that is searched.  Do you know anywhere that has a thorough explanation of this?

In case anyone else is interested...
I modified the code that you provided to do a MultiFieldQueryParse of the query and also to print the file path instead of name - the latter only returned null for me.

    /**
     * Search to return the document scores for a query
     * @param indexDir - the directory containing the index
     * @param queryStr - the query
     * @throws Exception
     */
    public void search(File indexDir, String queryStr)  throws Exception{
        Directory fsDir = FSDirectory.getDirectory(indexDir, false);
        IndexSearcher is = new IndexSearcher(fsDir);
       
        String[] fields = {"title",
                           "subject",
                           "creator",
                           "description",
                           "publisher",
                           "contributor",
                           "fullText"};
       
        Query query = MultiFieldQueryParser.parse(queryStr,
                                         fields,       
                                         new StandardAnalyzer());
        Hits hits = is.search(query);
        System.out.println("Found " + hits.length() +
                " document(s) that matched query '" + query + "':");
        for (int i = 0; i < hits.length(); i++) {
            Document doc = hits.doc(i);
           
            System.out.println("score : "+ hits.score(i));
            System.out.println(doc.get("path"));
        }
    }