<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Fraud-Detection on Thoughts and code</title>
    <link>https://claydon.co/tags/fraud-detection/</link>
    <description>Recent content in Fraud-Detection on Thoughts and code</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Wed, 01 Apr 2026 08:00:00 +0000</lastBuildDate><atom:link href="https://claydon.co/tags/fraud-detection/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Context Beats Magnitude: Why Embedding Sentences Outperforms Raw Numbers in Anomaly Detection</title>
      <link>https://claydon.co/code/its-just-vectors/part4-anomaly-detection/</link>
      <pubDate>Wed, 01 Apr 2026 08:00:00 +0000</pubDate>
      <guid>https://claydon.co/code/its-just-vectors/part4-anomaly-detection/</guid>
      <description>&lt;p&gt;&lt;code&gt;54.20&lt;/code&gt;. That&amp;rsquo;s what the transaction amount field contains. Embed that string and the model learns something about the number 54.20: that it sits near 54.19 and 54.21, that it&amp;rsquo;s a plausible price for groceries or a restaurant meal. It doesn&amp;rsquo;t learn anything about where the money went or when. This post uses a deliberately reductive fraud-shaped example to make a narrower point: formatting structured fields as a sentence gives the embedder more context than a raw scalar ever can.&lt;/p&gt;</description>
    </item>
    
  </channel>
</rss>
