Mailing-List: contact java-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: java-user@lucene.apache.org
Received-SPF: pass (athena.apache.org: domain of buschmic@gmail.com designates
 209.85.132.243 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=beta;
        h=received:message-id:date:from:user-agent:mime-version:to:subject:references:in-reply-to:content-type:content-transfer-encoding;
        b=pJYrlr1y8aG/8tHXqfBemEHKm7CMSD/goEG97y2Mex2/NsquNnhhlPCbOwP5YaYDgAhUANzVZ1UuWbYnpOFBgQi1WdlEDMG8bLZKGfWYcy8sLTnWDclTq0Ign1qYn6XitXPyzADFhE6X9YFod5MS6zrg+2MTSj0UvGn7vB4aqMA=
Message-ID: <473C9108.2080504@gmail.com>
Date: Thu, 15 Nov 2007 10:33:44 -0800
From: Michael Busch <buschmic@gmail.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: java-user@lucene.apache.org
Subject: Re: =?GB2312?B?tPC4tDogaG93IHRvIGVmZmVjaWVudGx5IGltcGxlbWVudCB0aA==?=
 =?GB2312?B?ZSBzdGFzdGljYWwgc2NvcmVzIGxpa2UgcGFnZXJhbms/?=
References: <000c01c8273e$00e7dc70$02b79550$@edu.cn>
	 <99dfded80711142222u30616dp6ef9282e047da98e@mail.gmail.com>
	 <000d01c8279b$8bf4f5c0$a3dee140$@edu.cn>
 <8837fb770711150703j6ec4c578qe829f74e5c61321b@mail.gmail.com>
In-Reply-To: <8837fb770711150703j6ec4c578qe829f74e5c61321b@mail.gmail.com>
Content-Type: text/plain; charset=GB2312
Content-Transfer-Encoding: 7bit

John Wang wrote:
> Would payload work?
> -John
>
>   
Yes, if you used payloads instead of stored fields your performance
should be much better.

Try and index one special term per document (e. g. score:pagerank), and
index one position with a payload for each doc. Then when you retrieve
hits open a TermPositions using the special term, get the payload and
incorporate it in the docs' score.

The performance overhead should be comparable to adding one AND-term to
your query.

-Michael

> On 11/15/07, Zhou Qi <zhou_qi@sjtu.edu.cn> wrote:
>   
>> Thank you, my score is fixed score from the properties of the page, but at
>> first we need to adjust the score for a promising result.
>> I have tried one way of manually re-ranking all the documents by the
>> search results. But it needs to iterate all the retrieved results and fetch
>> the re-ranking score (stored in the index) to sum the overall score. It is
>> inefficient. How to improve that by a new approach?
>>
>> Sorry for making you misunderstanding. Thanks
>>
>> Best Regards,
>>
>> Zhou Qi
>> Dept. Computer Science & Engineering
>> Shanghai Jiaotong University
>>
>>     


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