<?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: Identifying Places</title>
	<atom:link href="http://www.investigations.4-lom.com/2008/04/08/identifying-places/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.investigations.4-lom.com/2008/04/08/identifying-places/</link>
	<description>Failing better at understanding the past</description>
	<lastBuildDate>Mon, 12 Dec 2011 17:50:48 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: Gavin Robinson</title>
		<link>http://www.investigations.4-lom.com/2008/04/08/identifying-places/comment-page-1/#comment-13903</link>
		<dc:creator>Gavin Robinson</dc:creator>
		<pubDate>Fri, 11 Apr 2008 12:50:04 +0000</pubDate>
		<guid isPermaLink="false">http://www.investigations.4-lom.com/2008/04/08/identifying-places/#comment-13903</guid>
		<description>I&#039;ll have another post on this when I&#039;ve tested it some more. My first impression of comparing with the whole country instead of one county is that it gets too many results, but I think I need to vary the threshold from case to case instead of one-size-fits-all. In practice the amount of help I need to identify places varies a lot. Some are so obvious and unique that I can spot them without any help, others are completely undecidable as there are 2 or 3 equally plausible possibilities, and there&#039;s a lot of variation in between.</description>
		<content:encoded><![CDATA[<p>I&#8217;ll have another post on this when I&#8217;ve tested it some more. My first impression of comparing with the whole country instead of one county is that it gets too many results, but I think I need to vary the threshold from case to case instead of one-size-fits-all. In practice the amount of help I need to identify places varies a lot. Some are so obvious and unique that I can spot them without any help, others are completely undecidable as there are 2 or 3 equally plausible possibilities, and there&#8217;s a lot of variation in between.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: William J Turkel</title>
		<link>http://www.investigations.4-lom.com/2008/04/08/identifying-places/comment-page-1/#comment-13901</link>
		<dc:creator>William J Turkel</dc:creator>
		<pubDate>Thu, 10 Apr 2008 23:06:32 +0000</pubDate>
		<guid isPermaLink="false">http://www.investigations.4-lom.com/2008/04/08/identifying-places/#comment-13901</guid>
		<description>This is really interesting, Gavin.  I looked into the details of &lt;a href=&quot;http://pydoc.org/2.1/difflib.html&quot; rel=&quot;nofollow&quot;&gt;difflib&lt;/a&gt;, expecting to find some standard string distance measure like &lt;a href=&quot;http://en.wikipedia.org/wiki/Levenshtein_distance&quot; rel=&quot;nofollow&quot;&gt;Levenshtein&lt;/a&gt;, but it is funkier than that.  I will have to explore this further.  Best, Bill.</description>
		<content:encoded><![CDATA[<p>This is really interesting, Gavin.  I looked into the details of <a href="http://pydoc.org/2.1/difflib.html" rel="nofollow">difflib</a>, expecting to find some standard string distance measure like <a href="http://en.wikipedia.org/wiki/Levenshtein_distance" rel="nofollow">Levenshtein</a>, but it is funkier than that.  I will have to explore this further.  Best, Bill.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Pamphlets and programming &#171; Mercurius Politicus</title>
		<link>http://www.investigations.4-lom.com/2008/04/08/identifying-places/comment-page-1/#comment-13899</link>
		<dc:creator>Pamphlets and programming &#171; Mercurius Politicus</dc:creator>
		<pubDate>Tue, 08 Apr 2008 18:43:14 +0000</pubDate>
		<guid isPermaLink="false">http://www.investigations.4-lom.com/2008/04/08/identifying-places/#comment-13899</guid>
		<description>[...] April 2008 by mercuriuspoliticus    Gavin Robinson put up an interesting post today about using XML and Python to organise and double-check lists of [...]</description>
		<content:encoded><![CDATA[<p>[...] April 2008 by mercuriuspoliticus    Gavin Robinson put up an interesting post today about using XML and Python to organise and double-check lists of [...]</p>
]]></content:encoded>
	</item>
</channel>
</rss>

