Mailing-List: contact solr-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: solr-user@lucene.apache.org
Received-SPF: pass (nike.apache.org: domain of iorixxx@yahoo.com designates
 98.138.91.159 as permitted sender)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
  s=s1024; d=yahoo.com;
  h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding;
  b=TpoCpSbDyOqaHYSi65YxOdZzYVZQMnZq5ZyNbMnL/1XhNMrzjCYWwUeDaY312EwWmZs1H6qGEXTK3L81Uttst0XSobt7AlFJzJhSrtjyK6aGWfTOVXKdlsuZADvWsXKR66pKk7jMLAxBT3moi4K98PD7/PEJXXr6DHcJVxJnB/w=;
References: 
 <CAA72waAEa5HKCCfA6XO64Jbe2vhA=+pvA_KXMcJrwYU53Nxwhg@mail.gmail.com>
Message-ID: <1396949801.85086.YahooMailNeo@web124701.mail.ne1.yahoo.com>
Date: Tue, 8 Apr 2014 02:36:41 -0700 (PDT)
From: Ahmet Arslan <iorixxx@yahoo.com>
Reply-To: Ahmet Arslan <iorixxx@yahoo.com>
Subject: Re: Strange relevance scoring
To: "solr-user@lucene.apache.org" <solr-user@lucene.apache.org>
In-Reply-To: 
 <CAA72waAEa5HKCCfA6XO64Jbe2vhA=+pvA_KXMcJrwYU53Nxwhg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Hi=A0Nielsen,=0A=0AThere is no special attention paid to first word. You ar=
e probably hitting length normalisation.=A0=0ALucene/Solr punishes long doc=
uments, favours short documents.=A0=0A(5 times appearing one) longer?=0A=0A=
=0A=0AOn Tuesday, April 8, 2014 12:03 PM, John Nielsen <jn@mcb.dk> wrote:=
=0AHi,=0A=0AWe are seeing a strange phenomenon with our Solr setup which I =
have been=0Aunable to answer.=0A=0AMy Google-fu is clearly not up to the ta=
sk, so I am trying here.=0A=0AIt appears that if i do a freetext search for=
 a single word, say "modellering"=0Aon a text field, the scoring is massive=
ly boosted if the first word of the=0Atext field is a hit.=0A=0AFor instanc=
e if there is only one occurrence of the word "modellering" in=0Athe text f=
ield and that occurrence is the first word of the text, then that=0Adocumen=
t gets a higher relevancy than if the word "modelling" occurs 5=0Atimes in =
the text and the first word of the text is any other word.=0A=0AIs this nor=
mal behavior? Is special attention paid to the first word in a=0Atext field=
? I would think that the latter case would get the highest score.=0A=0A=0A-=
- =0AMed venlig hilsen / Best regards=0A=0A*John Nielsen*=0AProgrammer=0A=
=0A=0A=0A*MCB A/S*=0AEnghaven 15=0ADK-7500 Holstebro=0A=0AKundeservice: +45=
 9610 2824=0Apost@mcb.dk=0Awww.mcb.dk=0A