<?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: Algorithms for learning association rules in Zeitgeist</title>
	<atom:link href="http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/feed/" rel="self" type="application/rss+xml" />
	<link>http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/</link>
	<description>I like potato chips</description>
	<lastBuildDate>Wed, 08 Sep 2010 23:18:49 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.1</generator>
	<item>
		<title>By: Kostya</title>
		<link>http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/comment-page-1/#comment-4524</link>
		<dc:creator>Kostya</dc:creator>
		<pubDate>Mon, 08 Mar 2010 16:02:11 +0000</pubDate>
		<guid isPermaLink="false">http://seilo.geekyogre.com/?p=1074#comment-4524</guid>
		<description>Hi Seif, nice job! 
Just wanted to mention that apriori is not the best choice, since it isn&#039;t particularly developed for stream data mining. In [1] you will find some recent approaches to stream rule mining. There is also some work in clustering analysis of stream data that might be interesting for the project.

Jiang, N. and Gruenwald, L. 2006. Research issues in data stream association rule mining. SIGMOD Rec. 35, 1 (Mar. 2006), 14-19. DOI= http://doi.acm.org/10.1145/1121995.1121998</description>
		<content:encoded><![CDATA[<p>Hi Seif, nice job!<br />
Just wanted to mention that apriori is not the best choice, since it isn&#8217;t particularly developed for stream data mining. In [1] you will find some recent approaches to stream rule mining. There is also some work in clustering analysis of stream data that might be interesting for the project.</p>
<p>Jiang, N. and Gruenwald, L. 2006. Research issues in data stream association rule mining. SIGMOD Rec. 35, 1 (Mar. 2006), 14-19. DOI= <a href="http://doi.acm.org/10.1145/1121995.1121998" rel="nofollow">http://doi.acm.org/10.1145/1121995.1121998</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Seif Lotfy</title>
		<link>http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/comment-page-1/#comment-4212</link>
		<dc:creator>Seif Lotfy</dc:creator>
		<pubDate>Mon, 15 Feb 2010 22:39:43 +0000</pubDate>
		<guid isPermaLink="false">http://seilo.geekyogre.com/?p=1074#comment-4212</guid>
		<description>Hey
If you want i can show you the code and u can help us find a better way of pruning! right now i prune by looking just for a minimum tolerance for the frequency of item and items in transactions. I think there can be a better way to do so :)</description>
		<content:encoded><![CDATA[<p>Hey<br />
If you want i can show you the code and u can help us find a better way of pruning! right now i prune by looking just for a minimum tolerance for the frequency of item and items in transactions. I think there can be a better way to do so <img src='http://seilo.geekyogre.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Marcusklaas</title>
		<link>http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/comment-page-1/#comment-4211</link>
		<dc:creator>Marcusklaas</dc:creator>
		<pubDate>Mon, 15 Feb 2010 21:40:22 +0000</pubDate>
		<guid isPermaLink="false">http://seilo.geekyogre.com/?p=1074#comment-4211</guid>
		<description>Yo sup bro.. this was a cool article ;-) I actually got it ^_^ But I still think this will take a long ass time if you got a proper database. But that&#039;s just because it&#039;s pretty much an NP-complete problem, right? I did some pruning last year when we built a computer AI for the orthello game. If I recall correctly it gave us like a 2-3 time speed up there ;-) But I can imagine that with this type of problem pruning can even be more rewarding, especially when working on the  data set u described. 

Peace out bro</description>
		<content:encoded><![CDATA[<p>Yo sup bro.. this was a cool article <img src='http://seilo.geekyogre.com/wp-includes/images/smilies/icon_wink.gif' alt=';-)' class='wp-smiley' />  I actually got it ^_^ But I still think this will take a long ass time if you got a proper database. But that&#8217;s just because it&#8217;s pretty much an NP-complete problem, right? I did some pruning last year when we built a computer AI for the orthello game. If I recall correctly it gave us like a 2-3 time speed up there <img src='http://seilo.geekyogre.com/wp-includes/images/smilies/icon_wink.gif' alt=';-)' class='wp-smiley' />  But I can imagine that with this type of problem pruning can even be more rewarding, especially when working on the  data set u described. </p>
<p>Peace out bro</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Seif Lotfy</title>
		<link>http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/comment-page-1/#comment-4206</link>
		<dc:creator>Seif Lotfy</dc:creator>
		<pubDate>Mon, 15 Feb 2010 15:45:24 +0000</pubDate>
		<guid isPermaLink="false">http://seilo.geekyogre.com/?p=1074#comment-4206</guid>
		<description>The bigger the timerange the lesser the errors.
We are going to use an adaptive bucket size estimation based on the density of events. I think this would help out alot :)</description>
		<content:encoded><![CDATA[<p>The bigger the timerange the lesser the errors.<br />
We are going to use an adaptive bucket size estimation based on the density of events. I think this would help out alot <img src='http://seilo.geekyogre.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Christophe</title>
		<link>http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/comment-page-1/#comment-4205</link>
		<dc:creator>Christophe</dc:creator>
		<pubDate>Mon, 15 Feb 2010 14:57:11 +0000</pubDate>
		<guid isPermaLink="false">http://seilo.geekyogre.com/?p=1074#comment-4205</guid>
		<description>I really like what you are doing with zeitgeist and certainly these kind of functionalities.
The supermarket idea is a nice data mining example but is not fully comparable with event streams.
Because you assume that a transaction occurs every 30 minutes it is possible that events of different tasks will occur in one transaction.
I believe a better approach is to use instead algorithms for frequent episodes.</description>
		<content:encoded><![CDATA[<p>I really like what you are doing with zeitgeist and certainly these kind of functionalities.<br />
The supermarket idea is a nice data mining example but is not fully comparable with event streams.<br />
Because you assume that a transaction occurs every 30 minutes it is possible that events of different tasks will occur in one transaction.<br />
I believe a better approach is to use instead algorithms for frequent episodes.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Martyn Russell</title>
		<link>http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/comment-page-1/#comment-4202</link>
		<dc:creator>Martyn Russell</dc:creator>
		<pubDate>Mon, 15 Feb 2010 11:47:07 +0000</pubDate>
		<guid isPermaLink="false">http://seilo.geekyogre.com/?p=1074#comment-4202</guid>
		<description>Seif, why are you linking codethink.co.uk for Tracker? Codethink is not the only company working on Tracker. Codeminded AND Lanedo have core developers there too.</description>
		<content:encoded><![CDATA[<p>Seif, why are you linking codethink.co.uk for Tracker? Codethink is not the only company working on Tracker. Codeminded AND Lanedo have core developers there too.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Benjamin Humphrey</title>
		<link>http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/comment-page-1/#comment-4201</link>
		<dc:creator>Benjamin Humphrey</dc:creator>
		<pubDate>Mon, 15 Feb 2010 09:04:58 +0000</pubDate>
		<guid isPermaLink="false">http://seilo.geekyogre.com/?p=1074#comment-4201</guid>
		<description>That is awesome! Well done!

This sort of thing is totally my definition of cool. Well, as cool as it gets in the Ubuntu world. Which is cool for me, but for most others probably incredibly boring haha</description>
		<content:encoded><![CDATA[<p>That is awesome! Well done!</p>
<p>This sort of thing is totally my definition of cool. Well, as cool as it gets in the Ubuntu world. Which is cool for me, but for most others probably incredibly boring haha</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: uberVU - social comments</title>
		<link>http://seilo.geekyogre.com/2010/02/algorithms-for-learning-association-rules-in-zeitgeist/comment-page-1/#comment-4200</link>
		<dc:creator>uberVU - social comments</dc:creator>
		<pubDate>Mon, 15 Feb 2010 04:24:46 +0000</pubDate>
		<guid isPermaLink="false">http://seilo.geekyogre.com/?p=1074#comment-4200</guid>
		<description>&lt;strong&gt;Social comments and analytics for this post...&lt;/strong&gt;

This post was mentioned on Twitter by seiflotfy: New blog post: Algorithms for learning association rules in Zeitgeist http://bit.ly/c9YQYo...</description>
		<content:encoded><![CDATA[<p><strong>Social comments and analytics for this post&#8230;</strong></p>
<p>This post was mentioned on Twitter by seiflotfy: New blog post: Algorithms for learning association rules in Zeitgeist <a href="http://bit.ly/c9YQYo.." rel="nofollow">http://bit.ly/c9YQYo..</a>.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
