ISO/IEC 5259-2:2024 Artificial intelligence - Data quality for analytics and machine learning - Part 2: Data quality measures

• Publisher: ISO
• Status: final
• Version: 1
• Certifiable: False
• Release Date: 2024-11
• Date Added: 2025-11-25
• Source URL: https://www.iso.org/standard/81860.html

---

Summary

This standard specifies a data quality model, a set of measurable data quality characteristics, and guidance on reporting data quality in the context of analytics and machine learning. It builds on earlier ISO/IEC standards and supports organizations in assessing and monitoring data quality for ML and analytics use.

---

Key Takeaways

• Part 2 operationalizes the concepts from Part 1 by shifting from definitions to measurement criteria, enabling objective evaluation of data sets.
• It provides the structure needed for organizations to compare data quality assessments consistently across projects, teams, and ML pipelines.
• The measures outlined in this part help organizations identify where data preparation or remediation is needed before model development or deployment.
• It establishes a common basis for data quality reporting, supporting transparency and auditability in ML workflows.

---

Additional Sources

• ISO/IEC 5259-2: Data Quality Framework for Compliant AI — overview of the standard

---

Tags

data-quality, analytics, machine-learning, governance, ml

---