<?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: How Breck approaches new projects in natural language processing</title>
	<atom:link href="http://lingpipe-blog.com/2009/03/08/how-breck-approaches-new-projects-in-natural-language-processing/feed/" rel="self" type="application/rss+xml" />
	<link>http://lingpipe-blog.com/2009/03/08/how-breck-approaches-new-projects-in-natural-language-processing/</link>
	<description>Natural Language Processing and Text Analytics</description>
	<lastBuildDate>Fri, 24 May 2013 22:27:31 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.com/</generator>
	<item>
		<title>By: Michael Niv</title>
		<link>http://lingpipe-blog.com/2009/03/08/how-breck-approaches-new-projects-in-natural-language-processing/#comment-4130</link>
		<dc:creator><![CDATA[Michael Niv]]></dc:creator>
		<pubDate>Mon, 09 Mar 2009 21:48:31 +0000</pubDate>
		<guid isPermaLink="false">http://lingpipe-blog.com/?p=872#comment-4130</guid>
		<description><![CDATA[Breck,

Great writeup.  See 
http://www.gagsandgoods.com/index.php?productID=336
for help in case you don&#039;t manage to get an annotated dataset.

Michael.]]></description>
		<content:encoded><![CDATA[<p>Breck,</p>
<p>Great writeup.  See<br />
<a href="http://www.gagsandgoods.com/index.php?productID=336" rel="nofollow">http://www.gagsandgoods.com/index.php?productID=336</a><br />
for help in case you don&#8217;t manage to get an annotated dataset.</p>
<p>Michael.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: lingpipe</title>
		<link>http://lingpipe-blog.com/2009/03/08/how-breck-approaches-new-projects-in-natural-language-processing/#comment-4129</link>
		<dc:creator><![CDATA[lingpipe]]></dc:creator>
		<pubDate>Mon, 09 Mar 2009 18:18:42 +0000</pubDate>
		<guid isPermaLink="false">http://lingpipe-blog.com/?p=872#comment-4129</guid>
		<description><![CDATA[Nice.  This is like &lt;a href=&quot;http://www.pragprog.com/the-pragmatic-programmer&quot; rel=&quot;nofollow&quot;&gt;&lt;i&gt;The Pragmatic Programmer&lt;/i&gt;&lt;/a&gt; for machine learning projects.

The one case I&#039;ve seen where you can do better than people is when people with relatively little training and experience have to do a very broad classification in a very time-constrained fashion.  The obvious case here is search, but you could argue that&#039;s not natural language enough.  A better example is spoken call routing at large call centers, which is sort of part search, part classification, especially at a site like a bank or retailer that has all sorts of products.

- Bob [hey, maybe we should get our own logins?]]]></description>
		<content:encoded><![CDATA[<p>Nice.  This is like <a href="http://www.pragprog.com/the-pragmatic-programmer" rel="nofollow"><i>The Pragmatic Programmer</i></a> for machine learning projects.</p>
<p>The one case I&#8217;ve seen where you can do better than people is when people with relatively little training and experience have to do a very broad classification in a very time-constrained fashion.  The obvious case here is search, but you could argue that&#8217;s not natural language enough.  A better example is spoken call routing at large call centers, which is sort of part search, part classification, especially at a site like a bank or retailer that has all sorts of products.</p>
<p>- Bob [hey, maybe we should get our own logins?]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jochen L. Leidner</title>
		<link>http://lingpipe-blog.com/2009/03/08/how-breck-approaches-new-projects-in-natural-language-processing/#comment-4125</link>
		<dc:creator><![CDATA[Jochen L. Leidner]]></dc:creator>
		<pubDate>Mon, 09 Mar 2009 14:53:56 +0000</pubDate>
		<guid isPermaLink="false">http://lingpipe-blog.com/?p=872#comment-4125</guid>
		<description><![CDATA[I totally agree, except personally, I prefer to put the evaluation in place even before the implementation so that it can be used for regression testing.]]></description>
		<content:encoded><![CDATA[<p>I totally agree, except personally, I prefer to put the evaluation in place even before the implementation so that it can be used for regression testing.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
