Artificial intelligence — Data quality for analytics and machine learning (ML) — Part 1: Overview, terminology, and examples

Scope

This document provides the landscape for understanding and associating of “Data quality for analytics and ML” series and guides the foundational concepts regarding data quality for analytics and artificial intelligence. It also describes associated technologies and examples (e.g. use cases and usage scenarios).

This document does not define a detailed process for a data lifecycle, data quality management or data quality assessment issues. Rather it defines the overarching concepts and terminologies for understanding the scope and overview of “Data quality for analytics and ML” series. ©ISO/IEC 2022. All rights reserved.

Purpose

Data quality is one of the crucial challenges for organizations to successfully implement big data and AI systems. Therefore, most organizations are working to improve data quality from data collection to data analysis and for data use in model training for machine learning etc. No matter how good the data analytics and/or AI model performance is, if low quality data is inputted, the result will not reliable, and if a service using a model trained with low quality data is operated incorrectly, it can be a direct threat to safety.

Currently, several types of data quality standards have already been developed, but in the field of data analytics and artificial intelligence, quality according to various data characteristics (e.g. big data characteristics) should be considered along with the lifecycle of data.

This standard covers the overall concepts and the scope of “Data quality for analytics and ML” series. This standard also provides examples of data quality in analytics and ML to help you understand this series of standards. ©ISO/IEC 2022. All rights reserved.

External Links