<?xml version="1.0" encoding="utf-8"?><!-- generator="wordpress/2.2.2" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Evolutionary Music Composition and CrowdSourcing</title>
	<link>http://claus.castelodelego.org/archives/538</link>
	<description>Claus' little corner on the net</description>
	<pubDate>Thu, 09 Sep 2010 05:55:33 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.2.2</generator>

	<item>
		<title>By: Carlos Azevedo</title>
		<link>http://claus.castelodelego.org/archives/538#comment-45159</link>
		<author>Carlos Azevedo</author>
		<pubDate>Fri, 05 Mar 2010 20:24:17 +0000</pubDate>
		<guid>http://claus.castelodelego.org/archives/538#comment-45159</guid>
		<description>this could be described as a multi-objective problem, with many users voting based on different preferences: one could cluster users based on a priori declared music genre preferences and take the evaluation of the different groups (e.g. rock, jazz, latin, classical) as conflicting objectives. Then, one would try to approximate the Pareto front on this dynamic objective space (as more users are added into the database). This would indeed be an interesting IEC application. Of course, many other subjective preferences could be asked such as whether users prefer calm or agitated songs and so on.

But then, one wouldn't even need to explicitly translate the subjective perception concepts towards music input to the system because the whole system could be make a self-organizing one, with a clustering algorithm grouping people with similar voting pattern.

I really loved this idea. Wanna open a start-up with me? :-)</description>
		<content:encoded><![CDATA[<p>this could be described as a multi-objective problem, with many users voting based on different preferences: one could cluster users based on a priori declared music genre preferences and take the evaluation of the different groups (e.g. rock, jazz, latin, classical) as conflicting objectives. Then, one would try to approximate the Pareto front on this dynamic objective space (as more users are added into the database). This would indeed be an interesting IEC application. Of course, many other subjective preferences could be asked such as whether users prefer calm or agitated songs and so on.</p>
<p>But then, one wouldn&#8217;t even need to explicitly translate the subjective perception concepts towards music input to the system because the whole system could be make a self-organizing one, with a clustering algorithm grouping people with similar voting pattern.</p>
<p>I really loved this idea. Wanna open a start-up with me? <img src='http://claus.castelodelego.org/wp-includes/images/smilies/icon_smile.gif' alt=':-)' class='wp-smiley' /></p>
]]></content:encoded>
	</item>
</channel>
</rss>
