<?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: News Topics as Social Network</title>
	<atom:link href="http://flowingdata.com/2010/02/26/news-topics-as-social-network/feed/" rel="self" type="application/rss+xml" />
	<link>http://flowingdata.com/2010/02/26/news-topics-as-social-network/</link>
	<description>Strength in Numbers</description>
	<lastBuildDate>Sat, 11 Feb 2012 07:35:20 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: Dreamz Of India &#8212; Blog &#8212; Live Sexcam Frauen</title>
		<link>http://flowingdata.com/2010/02/26/news-topics-as-social-network/#comment-39505</link>
		<dc:creator>Dreamz Of India &#8212; Blog &#8212; Live Sexcam Frauen</dc:creator>
		<pubDate>Mon, 01 Mar 2010 21:01:02 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=5575#comment-39505</guid>
		<description>[...] &lt;b&gt;News&lt;/b&gt; Topics as Social Network &#124; FlowingData [...]</description>
		<content:encoded><![CDATA[<p>[...] &lt;b&gt;News&lt;/b&gt; Topics as Social Network | FlowingData [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: News Browsing by Network Analysis &#124; FUTURISTIC GROOVES</title>
		<link>http://flowingdata.com/2010/02/26/news-topics-as-social-network/#comment-39471</link>
		<dc:creator>News Browsing by Network Analysis &#124; FUTURISTIC GROOVES</dc:creator>
		<pubDate>Sun, 28 Feb 2010 22:25:16 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=5575#comment-39471</guid>
		<description>[...] Granted, the interface is currently hideous. But you wonder if something like this isn&#8217;t the future of news browsing. Can you imagine what happens when tagging technology gets truly semantic&#8211;when stories can be linked not just with keywords, but ideas? [Source: Flowing Data] [...]</description>
		<content:encoded><![CDATA[<p>[...] Granted, the interface is currently hideous. But you wonder if something like this isn&#8217;t the future of news browsing. Can you imagine what happens when tagging technology gets truly semantic&#8211;when stories can be linked not just with keywords, but ideas? [Source: Flowing Data] [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Han Kim</title>
		<link>http://flowingdata.com/2010/02/26/news-topics-as-social-network/#comment-39421</link>
		<dc:creator>Han Kim</dc:creator>
		<pubDate>Fri, 26 Feb 2010 20:28:01 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=5575#comment-39421</guid>
		<description>I guess, it is a form of Multi-Dimensional Scaling (MDS) to convert a data matrix to a data table with a reduced number of variables, as the PCA does. It is also called spatialization which make a map out of a non-geographic dataset.</description>
		<content:encoded><![CDATA[<p>I guess, it is a form of Multi-Dimensional Scaling (MDS) to convert a data matrix to a data table with a reduced number of variables, as the PCA does. It is also called spatialization which make a map out of a non-geographic dataset.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Kerim Can</title>
		<link>http://flowingdata.com/2010/02/26/news-topics-as-social-network/#comment-39408</link>
		<dc:creator>Kerim Can</dc:creator>
		<pubDate>Fri, 26 Feb 2010 16:45:43 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=5575#comment-39408</guid>
		<description>I have been reading about network graphs lately and arbitrary placement of nodes (and therefore arbitrary distances) for the sake of readability are encouraged. Having played with these graphs a little I think it is the right call. 

IMHO, a better way to represent strength is to create a dichotomous &quot;connections strength&quot; variable and use that to create &quot;transparent/solid&quot; links. Use the histogram of connections to decide where a reasonable cutoff is. For example, &quot;more than 5 connections&quot; is 1 and &quot;less than 5 connections&quot; is 0. Then make the connections that are 0 transparent (or light blue). Keep the connections that have the value of 1 solid (or dark blue).</description>
		<content:encoded><![CDATA[<p>I have been reading about network graphs lately and arbitrary placement of nodes (and therefore arbitrary distances) for the sake of readability are encouraged. Having played with these graphs a little I think it is the right call. </p>
<p>IMHO, a better way to represent strength is to create a dichotomous &#8220;connections strength&#8221; variable and use that to create &#8220;transparent/solid&#8221; links. Use the histogram of connections to decide where a reasonable cutoff is. For example, &#8220;more than 5 connections&#8221; is 1 and &#8220;less than 5 connections&#8221; is 0. Then make the connections that are 0 transparent (or light blue). Keep the connections that have the value of 1 solid (or dark blue).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Nathan Yau</title>
		<link>http://flowingdata.com/2010/02/26/news-topics-as-social-network/#comment-39406</link>
		<dc:creator>Nathan Yau</dc:creator>
		<pubDate>Fri, 26 Feb 2010 15:51:17 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=5575#comment-39406</guid>
		<description>i guess most of the time, yeah, but sometimes distances represents something like similarity or strength of relationship.</description>
		<content:encoded><![CDATA[<p>i guess most of the time, yeah, but sometimes distances represents something like similarity or strength of relationship.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Guillaume Daudin</title>
		<link>http://flowingdata.com/2010/02/26/news-topics-as-social-network/#comment-39402</link>
		<dc:creator>Guillaume Daudin</dc:creator>
		<pubDate>Fri, 26 Feb 2010 15:35:54 +0000</pubDate>
		<guid isPermaLink="false">http://flowingdata.com/?p=5575#comment-39402</guid>
		<description>I suspect distance is somewhat arbitrary and is about readability. Isn&#039;t that the way all these network-viewing software work ? (I might be very wrong, of course).</description>
		<content:encoded><![CDATA[<p>I suspect distance is somewhat arbitrary and is about readability. Isn&#8217;t that the way all these network-viewing software work ? (I might be very wrong, of course).</p>
]]></content:encoded>
	</item>
</channel>
</rss>

