<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: A Bird in the Hand&#8230;</title>
	<atom:link href="http://irgupf.com/2009/08/24/a-bird-in-the-hand/feed/" rel="self" type="application/rss+xml" />
	<link>http://irgupf.com/2009/08/24/a-bird-in-the-hand/</link>
	<description>Information Retrieval Research, Issues, and Discussion</description>
	<lastBuildDate>Tue, 14 Jun 2011 14:26:42 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.1.3</generator>
	<item>
		<title>By: jeremy</title>
		<link>http://irgupf.com/2009/08/24/a-bird-in-the-hand/comment-page-1/#comment-4132</link>
		<dc:creator>jeremy</dc:creator>
		<pubDate>Wed, 26 Aug 2009 05:40:51 +0000</pubDate>
		<guid isPermaLink="false">http://irgupf.com/?p=83#comment-4132</guid>
		<description>I hesitate to call Google&#039;s &quot;find similar&quot; a form of relevance feedback.  Rather, I would call it a form of &quot;query by document example&quot;.  

The reason I don&#039;t think that it is relevance feedback is that it doesn&#039;t actually rerank or augment your original search result list.  Rather, it issues a new query, and pulls back a completely new list.  

Relevance feedback (at least as far as I understand it) is more about augmentation and adjustment than wholesale replacement.  

But I agree with the second half of what you said; passage-level feedback, whether you&#039;re really doing feedback or doing a wholesale new query-by-passage query, is more useful than full-document feedback.</description>
		<content:encoded><![CDATA[<p>I hesitate to call Google&#8217;s &#8220;find similar&#8221; a form of relevance feedback.  Rather, I would call it a form of &#8220;query by document example&#8221;.  </p>
<p>The reason I don&#8217;t think that it is relevance feedback is that it doesn&#8217;t actually rerank or augment your original search result list.  Rather, it issues a new query, and pulls back a completely new list.  </p>
<p>Relevance feedback (at least as far as I understand it) is more about augmentation and adjustment than wholesale replacement.  </p>
<p>But I agree with the second half of what you said; passage-level feedback, whether you&#8217;re really doing feedback or doing a wholesale new query-by-passage query, is more useful than full-document feedback.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Gene Golovchinsky</title>
		<link>http://irgupf.com/2009/08/24/a-bird-in-the-hand/comment-page-1/#comment-4127</link>
		<dc:creator>Gene Golovchinsky</dc:creator>
		<pubDate>Wed, 26 Aug 2009 01:00:07 +0000</pubDate>
		<guid isPermaLink="false">http://irgupf.com/?p=83#comment-4127</guid>
		<description>One form of relevance feedback available on Google is the &#039;similar&#039; link that is found next to some (but not all?) search results. The problem is that it does one-shot, document-level relevance feedback, which means that the system doesn&#039;t really know what aspects of the document you found useful. Passage-level feedback would serve most people&#039;s needs better; more elaborate schemes that accrue evidence over multiple documents are difficult to get right due to &lt;a href=&quot;http://palblog.fxpal.com/?p=1398&quot; rel=&quot;nofollow&quot;&gt;hidden dependencies and viscosity&lt;/a&gt; associated with relevance feedback.</description>
		<content:encoded><![CDATA[<p>One form of relevance feedback available on Google is the &#8216;similar&#8217; link that is found next to some (but not all?) search results. The problem is that it does one-shot, document-level relevance feedback, which means that the system doesn&#8217;t really know what aspects of the document you found useful. Passage-level feedback would serve most people&#8217;s needs better; more elaborate schemes that accrue evidence over multiple documents are difficult to get right due to <a href="http://palblog.fxpal.com/?p=1398" rel="nofollow">hidden dependencies and viscosity</a> associated with relevance feedback.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

