<?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: A recommendation for Amazon.com’s recommendation system</title>
	<atom:link href="http://blog.cgsm.com/2010/03/02/a-recommendation-for-amazon-com%e2%80%99s-recommendation-system/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.cgsm.com/2010/03/02/a-recommendation-for-amazon-com%e2%80%99s-recommendation-system/</link>
	<description>Thoughts on Marketing and Other Stuff</description>
	<lastBuildDate>Sat, 28 Jan 2012 02:26:57 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: Thai Online Shopping</title>
		<link>http://blog.cgsm.com/2010/03/02/a-recommendation-for-amazon-com%e2%80%99s-recommendation-system/comment-page-1/#comment-948</link>
		<dc:creator>Thai Online Shopping</dc:creator>
		<pubDate>Wed, 21 Apr 2010 04:11:19 +0000</pubDate>
		<guid isPermaLink="false">http://blog.cgsm.com/?p=439#comment-948</guid>
		<description>Hi,
I&#039;ve been doing a lot of research online lately regarding traffic and SEO since I just opened up an Thai online shop with a couple of buddies. 
The information offered on your site would be of great value to GetSiam visitors, and I believe, your visitors would find great value in our site. Because our two websites are complementary rather than competitive, I see synergy here as an opportunity for our mutual benefit and I&#039;d really like to exchange links with your website. Have a look at our site if you want, we put a lot of work and effort into it and are dedicated to improve it constantly. 
We receive a lot of hits already (contact me for visitor stats) and I think both our sites can benefit from one another. 
Anyway, just thought I&#039;d share this with you - keep up the good work and get back to me if interested!</description>
		<content:encoded><![CDATA[<p>Hi,<br />
I&#8217;ve been doing a lot of research online lately regarding traffic and SEO since I just opened up an Thai online shop with a couple of buddies.<br />
The information offered on your site would be of great value to GetSiam visitors, and I believe, your visitors would find great value in our site. Because our two websites are complementary rather than competitive, I see synergy here as an opportunity for our mutual benefit and I&#8217;d really like to exchange links with your website. Have a look at our site if you want, we put a lot of work and effort into it and are dedicated to improve it constantly.<br />
We receive a lot of hits already (contact me for visitor stats) and I think both our sites can benefit from one another.<br />
Anyway, just thought I&#8217;d share this with you &#8211; keep up the good work and get back to me if interested!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Phoebe</title>
		<link>http://blog.cgsm.com/2010/03/02/a-recommendation-for-amazon-com%e2%80%99s-recommendation-system/comment-page-1/#comment-610</link>
		<dc:creator>Phoebe</dc:creator>
		<pubDate>Wed, 03 Mar 2010 12:36:54 +0000</pubDate>
		<guid isPermaLink="false">http://blog.cgsm.com/?p=439#comment-610</guid>
		<description>You have to credit Amazon for giving the option to mark &quot;Don&#039;t use for recommendations&quot; about any purchase, which I assume is meant to address the fact that people often buy gifts or make other purchases that don&#039;t fit with their overall preferences.

As you note, any recommendation system is only as effective as the data it has to leverage, and I think Amazon is limited by drawing only on correlations between users. Collaborative filtering has significant disadvantages as a recommendations approach, which I think is why we&#039;re seeing more recommendations systems taking other approaches - for example, at Jinni http://www.jinni.com our recommendations have a semantic basis.</description>
		<content:encoded><![CDATA[<p>You have to credit Amazon for giving the option to mark &#8220;Don&#8217;t use for recommendations&#8221; about any purchase, which I assume is meant to address the fact that people often buy gifts or make other purchases that don&#8217;t fit with their overall preferences.</p>
<p>As you note, any recommendation system is only as effective as the data it has to leverage, and I think Amazon is limited by drawing only on correlations between users. Collaborative filtering has significant disadvantages as a recommendations approach, which I think is why we&#8217;re seeing more recommendations systems taking other approaches &#8211; for example, at Jinni <a href="http://www.jinni.com" rel="nofollow">http://www.jinni.com</a> our recommendations have a semantic basis.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

