<?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: The Financial Times Rediscovers Text Mining</title>
	<atom:link href="http://arnoldit.com/wordpress/2008/10/11/the-financial-times-rediscovers-text-mining/feed/" rel="self" type="application/rss+xml" />
	<link>http://arnoldit.com/wordpress/2008/10/11/the-financial-times-rediscovers-text-mining/</link>
	<description>by Stephen E. Arnold</description>
	<lastBuildDate>Thu, 24 May 2012 08:08:47 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.2</generator>
	<item>
		<title>By: Ernest Perez</title>
		<link>http://arnoldit.com/wordpress/2008/10/11/the-financial-times-rediscovers-text-mining/comment-page-1/#comment-29174</link>
		<dc:creator>Ernest Perez</dc:creator>
		<pubDate>Sat, 15 Nov 2008 01:40:51 +0000</pubDate>
		<guid isPermaLink="false">http://arnoldit.com/wordpress/?p=2366#comment-29174</guid>
		<description>Hi, Steven,

Ran across your &quot;rediscovering text mining&quot; piece in the last few days. I was not familiar with the &quot;entity extraction&quot; usage, and enjoyed your explanation and dry humor.

My own background includes a Ph.D. in Library &amp; Information Studies. My specialty was designing &amp; managing news text retrieval systems (Houston Chronicle &amp; Chicago Sun-Times) and packages/portals giving end users complex combinations of resources &amp; services (Started &amp; managed the State Library of Oregon&#039;s online information system for State employees, plus developing taxonomy for the cluster of State websites, and implementing the search engine for State of Oregon Websites.

Since retiring, I&#039;ve consulted informally with Power Text Solutions, a company whose software technology I evaluated in a feature article for a 2003 issue of Online Magazine. They contacted me after retirement, and invited me to examine and comment on their &quot;next generation&quot; product.

I think you might enjoy taking a look at it. The core technology is &quot;iResearch Reporter.&quot;  Here&#039;s the text of a short description I wrote for one of my library publication contacts....

------------------------------------------------------
iResearch-Reporter by Power Text Solutions (PTS), of Salem OR, is tested software technology that is a &quot;virtual research assistant.&quot; It uses intelligent post-processing to select and add value to information from large multi-document sets retrieved via popular search engines.

PTS uses text-mining and summarization processing to analyze multiple documents sets. Then it uses linguistic analysis to create a readable, concise, accurate and organized summary report. The report contains inserted URL links throughout, for quick access to all source documents.

The end user gets instant intelligent analysis and summary text describing the major or important information topics about their query. It&#039;s great for:
    1) &quot;Quick study&quot; or comprehensive background information; as well as,
    2) Clear, concise, selected, organized and summarized information and links that serves as the entry to detailed study of the user&#039;s question area.

PTS uses any external computer retrieval system to identify groups of highly relevant documents relating to a search topic. iResearch-Reporter.com, the current demo system on the Web uses Google as the Search Engine. But in another installation, for a Homeland Security regional agency grouping in Texas, PTS software uses text-mining technology to search and summarize results from Google, Yahoo, and MSN, online databases, as well as local agency intelligence and security document collections.

The HLS system manager says, &quot;It can write a good summary report in a minute or two, where a human might have to scan maybe 15,000 documents to create an equivalent report.&quot;

PTS Technology on the Web:

iResearch Reporter offers a free Web demo research tool at . This has been the PTS  testbed site, with no active marketing up to this time. This site and preceding versions already serves thousands of users, mainly from Europe, with a growing U.S. .user base.

NewsFeed Researcher  PTS software technology produces detailed background reports of all news stories contained in Google NewsFeeds. It is throughout the day, essentially delivering complete, current background information about all major current news stories. [This one&#039;s impressive - total background information package on all current news stories!]

Both of the sites are available for free use. They&#039;re looking for specialized datafile information providers, for contract customization. In the meantime, PTS invites information professionals and the general public to try the systems out as a free demo.
------------------------------------------------------

This product is one of the best I&#039;ve seen. The text extracts and organized summaries are so good that you generally won&#039;t even notice that it was written by a computer system, rather than a human.

I really enjoyed your incisive analysis and judgement of the Trampoline product. In any case, I hope you may find this system of interest. As noted above, it&#039;s the best I&#039;ve seen in the text mining area.


Cheers,
  --ernest
_______________________
Ernest Perez, Ph.D.
1119 Satara Ct. NW
Salem OR 97304
503-588-3650 Home
503-884-4233 Cell
ernest.r.perez@gmail.com</description>
		<content:encoded><![CDATA[<p>Hi, Steven,</p>
<p>Ran across your &#8220;rediscovering text mining&#8221; piece in the last few days. I was not familiar with the &#8220;entity extraction&#8221; usage, and enjoyed your explanation and dry humor.</p>
<p>My own background includes a Ph.D. in Library &amp; Information Studies. My specialty was designing &amp; managing news text retrieval systems (Houston Chronicle &amp; Chicago Sun-Times) and packages/portals giving end users complex combinations of resources &amp; services (Started &amp; managed the State Library of Oregon&#8217;s online information system for State employees, plus developing taxonomy for the cluster of State websites, and implementing the search engine for State of Oregon Websites.</p>
<p>Since retiring, I&#8217;ve consulted informally with Power Text Solutions, a company whose software technology I evaluated in a feature article for a 2003 issue of Online Magazine. They contacted me after retirement, and invited me to examine and comment on their &#8220;next generation&#8221; product.</p>
<p>I think you might enjoy taking a look at it. The core technology is &#8220;iResearch Reporter.&#8221;  Here&#8217;s the text of a short description I wrote for one of my library publication contacts&#8230;.</p>
<p>&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;<br />
iResearch-Reporter by Power Text Solutions (PTS), of Salem OR, is tested software technology that is a &#8220;virtual research assistant.&#8221; It uses intelligent post-processing to select and add value to information from large multi-document sets retrieved via popular search engines.</p>
<p>PTS uses text-mining and summarization processing to analyze multiple documents sets. Then it uses linguistic analysis to create a readable, concise, accurate and organized summary report. The report contains inserted URL links throughout, for quick access to all source documents.</p>
<p>The end user gets instant intelligent analysis and summary text describing the major or important information topics about their query. It&#8217;s great for:<br />
    1) &#8220;Quick study&#8221; or comprehensive background information; as well as,<br />
    2) Clear, concise, selected, organized and summarized information and links that serves as the entry to detailed study of the user&#8217;s question area.</p>
<p>PTS uses any external computer retrieval system to identify groups of highly relevant documents relating to a search topic. iResearch-Reporter.com, the current demo system on the Web uses Google as the Search Engine. But in another installation, for a Homeland Security regional agency grouping in Texas, PTS software uses text-mining technology to search and summarize results from Google, Yahoo, and MSN, online databases, as well as local agency intelligence and security document collections.</p>
<p>The HLS system manager says, &#8220;It can write a good summary report in a minute or two, where a human might have to scan maybe 15,000 documents to create an equivalent report.&#8221;</p>
<p>PTS Technology on the Web:</p>
<p>iResearch Reporter offers a free Web demo research tool at . This has been the PTS  testbed site, with no active marketing up to this time. This site and preceding versions already serves thousands of users, mainly from Europe, with a growing U.S. .user base.</p>
<p>NewsFeed Researcher  PTS software technology produces detailed background reports of all news stories contained in Google NewsFeeds. It is throughout the day, essentially delivering complete, current background information about all major current news stories. [This one's impressive - total background information package on all current news stories!]</p>
<p>Both of the sites are available for free use. They&#8217;re looking for specialized datafile information providers, for contract customization. In the meantime, PTS invites information professionals and the general public to try the systems out as a free demo.<br />
&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;&#8212;</p>
<p>This product is one of the best I&#8217;ve seen. The text extracts and organized summaries are so good that you generally won&#8217;t even notice that it was written by a computer system, rather than a human.</p>
<p>I really enjoyed your incisive analysis and judgement of the Trampoline product. In any case, I hope you may find this system of interest. As noted above, it&#8217;s the best I&#8217;ve seen in the text mining area.</p>
<p>Cheers,<br />
  &#8211;ernest<br />
_______________________<br />
Ernest Perez, Ph.D.<br />
1119 Satara Ct. NW<br />
Salem OR 97304<br />
503-588-3650 Home<br />
503-884-4233 Cell<br />
<a href="mailto:ernest.r.perez@gmail.com">ernest.r.perez@gmail.com</a></p>
]]></content:encoded>
	</item>
</channel>
</rss>

