[IP] Google's search patent... (fwd)

From: radev@umich.edu
Date: Fri Feb 28 2003 - 17:04:39 EST

Next message: radev@umich.edu: "Untapped networks - about Duncan Watts"
Previous message: radev@umich.edu: "Weird things at Google"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Dave Farber wrote:
>From radev Fri Feb 28 17:03:52 2003
Return-Path: <listbox+trampoline+247+126316+706b33d5@v2.listbox.com>
Received: from imap.si.umich.edu
 by localhost with IMAP (fetchmail-5.2.3)
 for radev@localhost (single-drop); Fri, 28 Feb 2003 17:03:52 -0500 (EST)
Received: from krusty.si.umich.edu ([unix socket])
 by krusty.si.umich.edu (Cyrus v2.1.11) with LMTP; Fri, 28 Feb 2003 17:00:33 -0500
X-Sieve: CMU Sieve 2.2
Received: from pinkcadillac.mr.itd.umich.edu (pinkcadillac.mr.itd.umich.edu [141.211.14.45])
 by krusty.si.umich.edu (Postfix) with ESMTP id 7BC84C267
 for <radev@krusty.si.umich.edu>; Fri, 28 Feb 2003 17:00:33 -0500 (EST)
Received: (from daemon@localhost)
 by pinkcadillac.mr.itd.umich.edu (3.6u) with LDAP id h1SM3i502432
 for radev@mail.si.umich.edu; Fri, 28 Feb 2003 17:03:44 -0500 (EST)
Received: from lollipop.pobox.com (lollipop.listbox.com [208.210.125.29])
 by pinkcadillac.mr.itd.umich.edu (3.6u) with ESMTP id h1SM3iD02424
 for <radev@umich.edu>; Fri, 28 Feb 2003 17:03:44 -0500 (EST)
Received: by lollipop.pobox.com (Postfix, from userid 1001)
 id A82D12C69D5; Fri, 28 Feb 2003 17:03:42 -0500 (EST)
Received: from linc.cis.upenn.edu (LINC.CIS.UPENN.EDU [158.130.8.3])
 by umbrella.listbox.com (Postfix) with ESMTP id 1D0FC1680F6
 for <ip@v2.listbox.com>; Fri, 28 Feb 2003 17:01:15 -0500 (EST)
Received: from [169.229.7.54] (haas-wlan-44.AirBears.Berkeley.EDU [169.229.7.54])
 (authenticated bits=0)
 by linc.cis.upenn.edu (8.12.5/8.12.5) with ESMTP id h1SM1DkC022591
 for <ip@v2.listbox.com>; Fri, 28 Feb 2003 17:01:14 -0500 (EST)
User-Agent: Microsoft-Entourage/10.1.1.2418
Date: Fri, 28 Feb 2003 14:01:22 -0800
Subject: [IP] Google's search patent...
From: Dave Farber <dave@farber.net>
To: ip <ip@v2.listbox.com>
Message-ID: <BA851C32.33486%dave@farber.net>
Mime-version: 1.0
Content-type: multipart/alternative; boundary="B_3129285682_1734345"
Sender: owner-ip@v2.listbox.com
Precedence: list
Reply-To: dave@farber.net
List-ID: <ip@v2.listbox.com>
List-Help: <ip@v2.listbox.com">http://v2.listbox.com/help?list_name=ip@v2.listbox.com>
List-Subscribe: <mailto:subscribe-ip@v2.listbox.com>, <ip@v2.listbox.com">http://v2.listbox.com/subscribe/?listname=ip@v2.listbox.com>
List-Unsubscribe: <mailto:unsubscribe-ip@v2.listbox.com>, <ip@v2.listbox.com">http://v2.listbox.com/member/unsubscribe/?listname=ip@v2.listbox.com>
Errors-To: listbox+trampoline+247+126316+706b33d5@v2.listbox.com

> This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

--B_3129285682_1734345
Content-type: text/plain; charset="US-ASCII"
Content-transfer-encoding: 7bit

------ Forwarded Message
From: "Gillmor, Dan" <DGillmor@sjmercury.com>
Date: Fri, 28 Feb 2003 13:57:22 -0800
To: "'dave@farber.net'" <dave@farber.net>
Subject: Google's search patent...

news.com.com

Google lands Web search patent
By Stefanie Olsen
Staff Writer, CNET News.com
February 26, 2003, 5:02 PM PT
http://news.com.com/2100-1024-986204.html
Google this week was granted its first patent by the United States Patent
Office for a method of determining the relevance of Web pages in relation to
search queries.

The patent, which Google filed on Jan. 30, 2001, and was granted Tuesday,
governs methodology for parsing through Web documents to deliver Web surfers
the most relevant pages for their queries.

Specifically, it deals with "an improved search engine that refines a
document's relevance score based on interconnectivity of the document within
a set of relevant documents," according to a summary of the patent.

The invention could affect search companies that are building technology to
intelligently rank Web pages in relation to search queries. In the last
year, Web search has become one of the hottest markets on the Internet. Many
companies are furiously developing advanced tools and techniques that will
index the Web more effectively and so, they hope, draw visitors.

As the top destination site for online searches, Google fields more than 150
million worldwide queries every day. When a visitor types a keyword into the
search field, its Web servers send the request to an index server, which
identifies Web pages containing words that match the query. Document servers
with the matching pages deliver links to the visitor in less than half a
second, according to Google's site.

The new patent deals with the process for finding matching documents. Under
the methodology, Google turns up an initial set of documents related to the
keyword and then ranks each page with a "relevance score." Next, it
calculates a "local score value" that quantifies "an amount that the
documents are referenced by other documents in the generated set of
documents," according to the filing. Finally, the local score values
influence the relevance ranking of a page.

According to the patent, "a search engine modifies the relevance rankings
for a set of documents based on the interconnectivity of the documents in
the set. A document with a high interconnectivity with other documents in
the initial set of relevant documents indicates that the document has
'support' in the set, and the document's new ranking will increase. In this
manner, the search engine re-ranks the initial set of ranked documents to
thereby refine the initial rankings."

The invention's assignee is Mountain View, Calif.-based Google, and its
inventor is Krishna Bharat, a senior research scientist at the company. He
holds a doctorate in computer science from Georgia Institute of Technology.

Google would not comment on the patent.

The company now has three outstanding patent applications. Two concern
methods and technology for providing search results in response to an
ambiguous search query. The third deals with methodology and technology for
delivering search results that use analysis of Web page usage.

In addition, Google co-founder Larry Page invented a methodology called
PageRank, which was patented to the board of trustees of the Leland Stanford
Junior University on September 2001. PageRank is one of Google's recipes for
calculating the popularity and relevance of Web pages based on the number of
other pages linking to it.

-------------------------------------
You are subscribed as radev@umich.edu
To unsubscribe or update your address, click
http://v2.listbox.com/member/?listname=ip

Archives at: http://www.interesting-people.org/archives/interesting-people/
--B_3129285682_1734345
Content-type: text/html; charset="US-ASCII"
Content-transfer-encoding: quoted-printable

<HTML>
<HEAD>
<TITLE>approve:tippie  Google's search patent...</TITLE>
</HEAD>
<BODY>
 
------ Forwarded Message 
From: "Gillmor, Dan" <DGillmor@sjmercury.com> 
Date: Fri, 28 Feb 2003 13:57:22 -0800 
To: "'dave@farber.net'" <dave@farber.net> 
Subject: Google's search patent... 
 
news.com.com 
 
Google lands Web=
search patent 
By Stefanie Olsen 
Staff Writer, CNET News.com 
February 26, 2003, 5:02 PM PT 
http://news.com.com/2100-1024-986204.html></F=
ONT> 
Google this week was granted its first patent by the United States Paten=
t Office for a method of determining the relevance of Web pages in relation=
to search queries. 
 
The patent, which Google filed on Jan=
. 30, 2001, and was granted Tuesday, governs methodology for parsing throug=
h Web documents to deliver Web surfers the most relevant pages for their qu=
eries. 
 
Specifically, it deals with "an improved search engine that refines a =
document's relevance score based on interconnectivity of the document withi=
n a set of relevant documents," according to a summary of the patent. =
 
 
The invention could affect search companies that are building technology to=
intelligently rank Web pages in relation to search queries. In the last ye=
ar, Web search has become one of the hottest=
 markets on the Internet. Many companies are furiously developing ad=
vanced tools and techniques that will index the Web more effectively and so=
, they hope, draw visitors. 
 
As the top destination site for online searches, <U=
>Google fields more than 150 million worldwide queries every day=
. When a visitor types a keyword into the search field, its Web servers sen=
d the request to an index server, which identifies Web pages containing wor=
ds that match the query. Document servers with the matching pages deliver l=
inks to the visitor in less than half a second, according to Google's site.=
 
 
The new patent deals with the process for finding matching documents. Under=
the methodology, Google turns up an initial set of documents related to th=
e keyword and then ranks each page with a "relevance score." Next=
, it calculates a "local score value" that quantifies "an am=
ount that the documents are referenced by other documents in the generated =
set of documents," according to the filing. Finally, the local score v=
alues influence the relevance ranking of a page. 
 
According to the patent, "a search engine modifies the relevance ranki=
ngs for a set of documents based on the interconnectivity of the documents =
in the set. A document with a high interconnectivity with other documents i=
n the initial set of relevant documents indicates that the document has 'su=
pport' in the set, and the document's new ranking will increase. In this ma=
nner, the search engine re-ranks the initial set of ranked documents to the=
reby refine the initial rankings." 
 
The invention's assignee is Mountain View, Calif.-based Google, and its inv=
entor is Krishna Bharat, a senior res=
earch scientist at the company. He holds a doctorate in computer science fr=
om Georgia Institute of Technology. 
 
Google would not comment on the patent. 
 
The company now has three outstanding patent applications. Two concern meth=
ods and technology for providing search results in response to an ambiguous=
search query. The third deals with methodology and technology for deliveri=
ng search results that use analysis of Web page usage. 
 
In addition, Google co-founder Larry Page invented a methodology called Pag=
eRank, which was patented to the board of trustees of the Leland Stanford J=
unior University on September 2001. PageRank is one of Google's recipes for=
calculating the popularity and relevance of Web pages based on the number =
of other pages linking to it. 
 

</BODY>
</HTML>

--B_3129285682_1734345--

-- 
Dragomir R. Radev                                         radev@umich.edu
Assistant Professor of Information, Electrical Engineering and
Computer Science, and Linguistics, the University of Michigan, Ann Arbor
Phone: 734-615-5225   Fax: 734-764-2475    http://www.si.umich.edu/~radev

Next message: radev@umich.edu: "Untapped networks - about Duncan Watts"
Previous message: radev@umich.edu: "Weird things at Google"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b30 : Tue Jun 09 2009 - 05:00:07 EDT