ISO/IEC 5259 Artificial intelligence: Data quality for analytics and machine learning (ML)

The standard developed jointly by the International Organisation for Standardisation and the International Electrotechnical Commission, ISO/IEC 5259, provides a comprehensive framework for ensuring and managing data quality in analytics and machine learning (ML) environments. This standard establishes guidelines for defining, measuring, managing and controlling data quality throughout its lifecycle – from acquisition and preparation through use and evaluation – to ensure that analytical and artificial intelligence (AI)-based results are reliable, explainable and trustworthy.

ISO/IEC 5259 Artificial intelligence

Watch video

Executive summary

ISO/IEC 5259 promotes a structured approach to data quality management, aimed at helping organizations ensure reliable and transparent results when using AI. It provides a framework for defining, measuring and governing data quality, integrating it into existing governance, risk and compliance systems. It also promotes traceability, accountability and organizational resilience, encouraging the alignment of quality practices with the strategic and ethical objectives of each organization.

The standard has five parts, covering the principles, models and processes necessary for effective and consistent data quality management in analytics and ML.

Main content

• Overview, terminology and examples (Part 1). Introduces the principles, scope, and terminology of data-quality management in analytics and ML, providing illustrative examples to create a shared understanding across technical and governance teams.
• Data-quality measures (Part 2). Defines measurable characteristics such as accuracy, completeness, consistency, timeliness, and representativeness to assess and monitor the quality of data used in analytics and ML systems.
• Data-quality management requirements and guidance (Part 3). Establishes requirements and practical guidance for implementing and continuously improving data-quality management processes aligned with organizational objectives.
• Data-quality process framework (Part 4). Provides a standardized process model covering all phases of the data lifecycle, ensuring consistency and quality control in data preparation, evaluation, and deployment.
• Data-quality governance framework (Part 5). Defines governance structures, roles, and responsibilities to ensure oversight, accountability, and alignment of data-quality activities with strategic goals and regulatory expectations.

Download the technical note on the ISO/IEC 5259 Artificial intelligence: Data quality for analytics and machine learning (ML).