<?xml version="1.0" encoding="UTF-8"?>
<rss  xmlns:atom="http://www.w3.org/2005/Atom" 
      xmlns:media="http://search.yahoo.com/mrss/" 
      xmlns:content="http://purl.org/rss/1.0/modules/content/" 
      xmlns:dc="http://purl.org/dc/elements/1.1/" 
      version="2.0">
<channel>
<title>DataMetricus Insights</title>
<link>https://datametricus.com/insights.html</link>
<atom:link href="https://datametricus.com/insights.xml" rel="self" type="application/rss+xml"/>
<description>Practical notes on model validation, actuarial analytics, health metrics, and reproducible quantitative work.</description>
<language>en-GB</language>
<generator>quarto-1.8.27</generator>
<lastBuildDate>Fri, 23 Oct 2026 22:00:00 GMT</lastBuildDate>
<item>
  <title>Why 95% Accuracy Can Miss Every Positive Case</title>
  <dc:creator>Federica Gazzelloni</dc:creator>
  <link>https://datametricus.com/insights/accuracy-can-miss-every-positive-case.html</link>
  <description><![CDATA[ A practical explanation of why accuracy alone can hide model failure, and what to examine before relying on a classification result. ]]></description>
  <category>Model validation</category>
  <category>Classification</category>
  <category>Decision metrics</category>
  <guid>https://datametricus.com/insights/accuracy-can-miss-every-positive-case.html</guid>
  <pubDate>Fri, 23 Oct 2026 22:00:00 GMT</pubDate>
</item>
</channel>
</rss>
