<?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: What makes recommender systems work?</title>
	<atom:link href="http://lemire.me/blog/archives/2008/12/31/what-makes-recommender-systems-work/feed/" rel="self" type="application/rss+xml" />
	<link>http://lemire.me/blog/archives/2008/12/31/what-makes-recommender-systems-work/</link>
	<description>Computer Scientist and Open Scholar: Databases, Information Retrieval, Business Intelligence.</description>
	<lastBuildDate>Thu, 24 May 2012 04:06:27 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.2</generator>
	<item>
		<title>By: marcel</title>
		<link>http://lemire.me/blog/archives/2008/12/31/what-makes-recommender-systems-work/comment-page-1/#comment-50513</link>
		<dc:creator>marcel</dc:creator>
		<pubDate>Wed, 07 Jan 2009 09:30:33 +0000</pubDate>
		<guid isPermaLink="false">http://www.daniel-lemire.com/blog/?p=1700#comment-50513</guid>
		<description>I disagree in the last paragraph. Diversity is essential to predict tastes. Having a &quot;delta function&quot; in the data (ratings -&gt; correlation etc.) means no information, hence no prediction power. A broad distribution of tastes always leads to better results. To predict blockbusters is an easy task, but not very useful. I claim: collaborative filtering works because of divergent tastes. The more divergent the better.</description>
		<content:encoded><![CDATA[<p>I disagree in the last paragraph. Diversity is essential to predict tastes. Having a &#8220;delta function&#8221; in the data (ratings -&gt; correlation etc.) means no information, hence no prediction power. A broad distribution of tastes always leads to better results. To predict blockbusters is an easy task, but not very useful. I claim: collaborative filtering works because of divergent tastes. The more divergent the better.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Daniel Tunkelang</title>
		<link>http://lemire.me/blog/archives/2008/12/31/what-makes-recommender-systems-work/comment-page-1/#comment-50470</link>
		<dc:creator>Daniel Tunkelang</dc:creator>
		<pubDate>Wed, 31 Dec 2008 15:05:33 +0000</pubDate>
		<guid isPermaLink="false">http://www.daniel-lemire.com/blog/?p=1700#comment-50470</guid>
		<description>Nice, I didn&#039;t realize you were a fan of the heuristics and biases literature. I recommend you read George Loewenstein and Dan Ariely if you haven&#039;t already.

http://sds.hss.cmu.edu/src/faculty/loewenstein.php

http://web.mit.edu/ariely/www/MIT/papers.shtml</description>
		<content:encoded><![CDATA[<p>Nice, I didn&#8217;t realize you were a fan of the heuristics and biases literature. I recommend you read George Loewenstein and Dan Ariely if you haven&#8217;t already.</p>
<p><a href="http://sds.hss.cmu.edu/src/faculty/loewenstein.php" rel="nofollow">http://sds.hss.cmu.edu/src/faculty/loewenstein.php</a></p>
<p><a href="http://web.mit.edu/ariely/www/MIT/papers.shtml" rel="nofollow">http://web.mit.edu/ariely/www/MIT/papers.shtml</a></p>
]]></content:encoded>
	</item>
</channel>
</rss>

