<?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/"
	xmlns:georss="http://www.georss.org/georss" xmlns:geo="http://www.w3.org/2003/01/geo/wgs84_pos#" xmlns:media="http://search.yahoo.com/mrss/"
		>
<channel>
	<title>Comments on: If you liked my blog, you&#8217;d like this post. Trust me!</title>
	<atom:link href="http://alteregozi.com/2008/11/25/if-you-liked-my-blog-youd-like-this-post-trust-me/feed/" rel="self" type="application/rss+xml" />
	<link>http://alteregozi.com/2008/11/25/if-you-liked-my-blog-youd-like-this-post-trust-me/</link>
	<description></description>
	<lastBuildDate>Mon, 20 Jun 2011 12:25:47 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.com/</generator>
	<item>
		<title>By: Farewell Academia &#124; The Alter Egozi</title>
		<link>http://alteregozi.com/2008/11/25/if-you-liked-my-blog-youd-like-this-post-trust-me/#comment-352</link>
		<dc:creator><![CDATA[Farewell Academia &#124; The Alter Egozi]]></dc:creator>
		<pubDate>Sat, 14 May 2011 18:01:04 +0000</pubDate>
		<guid isPermaLink="false">http://alteregozi.wordpress.com/?p=120#comment-352</guid>
		<description><![CDATA[[...] due to a statistically significant increment of 5% in relevancy (see the major interest around the Netflix prize), whereas actual users will barely feel the difference. On the other hand, stuff that is considered [...]]]></description>
		<content:encoded><![CDATA[<p>[...] due to a statistically significant increment of 5% in relevancy (see the major interest around the Netflix prize), whereas actual users will barely feel the difference. On the other hand, stuff that is considered [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Friendly advice from your &#8220;Social Trust Graph&#8221; &#171; The Alter Egozi</title>
		<link>http://alteregozi.com/2008/11/25/if-you-liked-my-blog-youd-like-this-post-trust-me/#comment-170</link>
		<dc:creator><![CDATA[Friendly advice from your &#8220;Social Trust Graph&#8221; &#171; The Alter Egozi]]></dc:creator>
		<pubDate>Tue, 28 Jul 2009 19:39:06 +0000</pubDate>
		<guid isPermaLink="false">http://alteregozi.wordpress.com/?p=120#comment-170</guid>
		<description><![CDATA[[...] the netflix prize winning team approach (see previous post) &#8220;prove&#8221; that statistical analysis of the user-item matrix beats any external signal [...]]]></description>
		<content:encoded><![CDATA[<p>[...] the netflix prize winning team approach (see previous post) &#8220;prove&#8221; that statistical analysis of the user-item matrix beats any external signal [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ofer Egozi</title>
		<link>http://alteregozi.com/2008/11/25/if-you-liked-my-blog-youd-like-this-post-trust-me/#comment-15</link>
		<dc:creator><![CDATA[Ofer Egozi]]></dc:creator>
		<pubDate>Wed, 26 Nov 2008 07:34:38 +0000</pubDate>
		<guid isPermaLink="false">http://alteregozi.wordpress.com/?p=120#comment-15</guid>
		<description><![CDATA[Well, I wouldn&#039;t say it&#039;s not AI. A lot of AI research follows these lines, and the state of the art in machine translation (which is NLP, therefore AI) for example, is a 100% statistical approach, with almost no expert knowledge or reasoning involved. And the crowds intelligence sub-field even has a fancy name, it&#039;s &lt;a href=&quot;http://en.wikipedia.org/wiki/Collaborative_filtering&quot; rel=&quot;nofollow&quot;&gt;collaborative filtering&lt;/a&gt;...
But you are right that &lt;a href=&quot;http://en.wikipedia.org/wiki/Expert_system&quot; rel=&quot;nofollow&quot;&gt;expert-systems&lt;/a&gt; have lost their appeal  as an AI tool, in favor of statistical approaches, and that leads to less reasoning possible over results.
Napoleon Dynamite - you&#039;re right, he also mentioned that briefly. That probably is also why &quot;Miss Congeniality&quot; is the most rated movie in the dataset... :-)]]></description>
		<content:encoded><![CDATA[<p>Well, I wouldn&#8217;t say it&#8217;s not AI. A lot of AI research follows these lines, and the state of the art in machine translation (which is NLP, therefore AI) for example, is a 100% statistical approach, with almost no expert knowledge or reasoning involved. And the crowds intelligence sub-field even has a fancy name, it&#8217;s <a href="http://en.wikipedia.org/wiki/Collaborative_filtering" rel="nofollow">collaborative filtering</a>&#8230;<br />
But you are right that <a href="http://en.wikipedia.org/wiki/Expert_system" rel="nofollow">expert-systems</a> have lost their appeal  as an AI tool, in favor of statistical approaches, and that leads to less reasoning possible over results.<br />
Napoleon Dynamite &#8211; you&#8217;re right, he also mentioned that briefly. That probably is also why &#8220;Miss Congeniality&#8221; is the most rated movie in the dataset&#8230; <img src='http://s0.wp.com/wp-includes/images/smilies/icon_smile.gif' alt=':-)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Pasha</title>
		<link>http://alteregozi.com/2008/11/25/if-you-liked-my-blog-youd-like-this-post-trust-me/#comment-14</link>
		<dc:creator><![CDATA[Pasha]]></dc:creator>
		<pubDate>Tue, 25 Nov 2008 22:03:11 +0000</pubDate>
		<guid isPermaLink="false">http://alteregozi.wordpress.com/?p=120#comment-14</guid>
		<description><![CDATA[&quot;Not being able to intelligently reason over the underlying logic exposed by an AI software is a tremendous drawback in my eyes&quot;

That&#039;s because this is not AI at all. This is just &quot;crowds intelligence&quot; software. I think what is really saddening you is the fact that after so many years of trying, AI approaches are still less successful than others.

Another curious anecdote related to the NetFlix challenge is the &quot;Napoleon Dynamite&quot; problem (http://www.nytimes.com/2008/11/23/magazine/23Netflix-t.html?pagewanted=all). Some movies are very &quot;polarizing&quot; and so, most NN approaches tend to struggle with them.]]></description>
		<content:encoded><![CDATA[<p>&#8220;Not being able to intelligently reason over the underlying logic exposed by an AI software is a tremendous drawback in my eyes&#8221;</p>
<p>That&#8217;s because this is not AI at all. This is just &#8220;crowds intelligence&#8221; software. I think what is really saddening you is the fact that after so many years of trying, AI approaches are still less successful than others.</p>
<p>Another curious anecdote related to the NetFlix challenge is the &#8220;Napoleon Dynamite&#8221; problem (<a href="http://www.nytimes.com/2008/11/23/magazine/23Netflix-t.html?pagewanted=all" rel="nofollow">http://www.nytimes.com/2008/11/23/magazine/23Netflix-t.html?pagewanted=all</a>). Some movies are very &#8220;polarizing&#8221; and so, most NN approaches tend to struggle with them.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

