<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Glossary on</title><link>https://docs.protegrity.com/synthetic-data/1.0.1/glossary/</link><description>Recent content in Glossary on</description><generator>Hugo</generator><language>en</language><atom:link href="https://docs.protegrity.com/synthetic-data/1.0.1/glossary/index.xml" rel="self" type="application/rss+xml"/><item><title>Metrics Knowledge Base</title><link>https://docs.protegrity.com/synthetic-data/1.0.1/glossary/metrics_knowledge_base/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://docs.protegrity.com/synthetic-data/1.0.1/glossary/metrics_knowledge_base/</guid><description>&lt;h2 id="propensity-score-mean-squared-error-pmse">Propensity Score Mean Squared Error (PMSE)&lt;/h2>
&lt;p>Measures how well synthetic data mimics real data by testing whether a machine learning classifier can distinguish between them.&lt;/p>
&lt;p>&lt;strong>How it works:&lt;/strong> Combines real and synthetic data with labels, trains a Random Forest classifier using cross-validation, and calculates the squared difference between predicted probabilities and actual proportions.&lt;/p>
&lt;p>&lt;strong>Interpretation:&lt;/strong>&lt;/p>
&lt;ul>
&lt;li>0.0 → Perfect similarity, classifier cannot distinguish synthetic from real data&lt;/li>
&lt;li>
&lt;blockquote>
&lt;p>0.25 → Poor similarity, classifier easily detects synthetic data&lt;/p></description></item><item><title>Protegrity-specific term definitions</title><link>https://docs.protegrity.com/synthetic-data/1.0.1/glossary/glossary/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://docs.protegrity.com/synthetic-data/1.0.1/glossary/glossary/</guid><description>&lt;h2 id="application-data-security">Application Data Security&lt;/h2>
&lt;p>Application data security is the process of protecting sensitive data used within an application during processing, storage, or transmission. Security measures such as encryption, tokenization, or data masking are applied to ensure that sensitive information remains secure from unauthorized access or data breaches. Application data security is important for ensuring that personal or financial information is protected in enterprise applications.&lt;/p>
&lt;h2 id="application-programming-interface-security">Application Programming Interface Security&lt;/h2>
&lt;p>Application programming interface security refers to the practice of protecting an Application Programming Interface (API) from unauthorized access, data breaches, or misuse. This involves applying security measures such as authentication, encryption, and access control. These measures secure the interactions between systems that communicate through APIs. API security is essential in modern applications where APIs are used to integrate services across cloud environments.&lt;/p></description></item></channel></rss>