ISO IEC 5259-3-2024 PDF

Full title and description

ISO/IEC 5259-3:2024 — Artificial intelligence — Data quality for analytics and machine learning (ML) — Part 3: Data quality management requirements and guidelines. This International Standard specifies requirements and provides guidance for establishing, implementing, maintaining and continually improving the quality of data used for analytics and machine learning applications.

Abstract

Part 3 of the ISO/IEC 5259 series defines a flexible data quality management framework (DQMS) for analytics and ML. It sets out high‑level requirements and guidance for planning, governance, roles and responsibilities, traceability and continual improvement of data quality across the data life cycle. The document is intentionally generic: it does not prescribe specific metrics, tools or a single process model but instead provides requirements that organisations can tailor to their context and AI life cycle.

General information

• Status: Published (International Standard).
• Publication date: July 2024.
• Publisher: Joint ISO/IEC standard (prepared under ISO/IEC JTC 1/SC 42) and made available by ISO and national standards bodies.
• ICS / categories: 35.020 (Information technology — IT in general).
• Edition / version: Edition 1 (2024).
• Number of pages: 28 pages (ISO published edition; national reprints or national adoption documents may include additional forewords/annex material).

Scope

This standard applies to organisations that create, collect, process or use data for analytics and machine learning and that need a structured, auditable approach to data quality. It covers requirements and guidance for a data quality management system that can be integrated with other management systems (for example quality, security or AI management systems) and adapted to different AI life cycles. The document is applicable to organisations of any size or sector and is deliberately generic to allow tailoring.

Key topics and requirements

• Establishing and maintaining a Data Quality Management System (DQMS) tailored to analytics and ML requirements.
• Data life‑cycle governance: acquisition, labelling, processing, storage, use, retention and disposal with traceability and provenance.
• Roles, responsibilities and accountability for data quality across organisational functions.
• Risk‑based approach to data quality, including identification of data quality risks that could affect ML outcomes.
• Data quality dimensions and attributes to consider (accuracy, completeness, consistency, timeliness, representativeness, lineage and fitness for purpose).
• Requirements for monitoring, measurement, auditability and continual improvement of data quality practices.
• Guidance for integrating DQMS with related management standards and for tailoring methods and metrics to organisational context.

Typical use and users

Intended users include data governance teams, data engineers, ML engineers, AI program managers, compliance and audit teams, risk officers, and organisations seeking to demonstrate auditable, repeatable data quality management for analytics/ML. Certification bodies and auditors may use the document as a normative reference when assessing data quality processes for AI systems.

Related standards

ISO/IEC 5259-3:2024 is part of the ISO/IEC 5259 series on data quality for analytics and ML and is intended to be used alongside other AI and management standards such as ISO/IEC 42001 (AI management systems), ISO 9001 (quality management), and ISO/IEC 27001 (information security). National adoptions or identical adoptions (for example CSA/ANSI listings) may exist.

Keywords

data quality, data quality management system (DQMS), analytics, machine learning, AI governance, data lineage, traceability, data life cycle, ISO/IEC 5259, JTC 1/SC 42.

FAQ

Q: What is this standard?

A: ISO/IEC 5259-3:2024 is an international standard that specifies requirements and provides guidance for establishing and running a data quality management system for analytics and machine learning.

Q: What does it cover?

A: It covers high‑level requirements and guidance for planning, governance, roles and responsibilities, traceability, monitoring and continual improvement of data quality across the data life cycle used in analytics and ML. It does not mandate specific metrics or detailed processes, leaving those choices to organisations to tailor to their context.

Q: Who typically uses it?

A: Data governance teams, data and ML engineers, compliance and audit professionals, AI programme managers, risk officers and organisations building, maintaining or certifying analytics/ML systems. Certification and assurance bodies may reference the standard in audits and assessments.

Q: Is it current or superseded?

A: Current — first edition published in July 2024. Organisations should check national bodies for adopted or amended national versions.

Q: Is it part of a series?

A: Yes — it is Part 3 of the ISO/IEC 5259 series addressing data quality for analytics and ML; other parts of the series address complementary topics and guidance.

Q: What are the key keywords?

A: Data quality, DQMS, analytics, machine learning, traceability, provenance, governance, continual improvement.