Multidimensional Quality Metrics (MQM) is a framework for analytic Translation Quality Evaluation (TQE). It can be applied to both human translation, machine translation and AI-generated translation.
As part of a quality management system, MQM can be used to identify quality issues in translation products, classify them against a shared, open and standardized error typology, and generate quality measures that can be used to gauge how well the translation product meets quality requirements.
The standardized MQM approach generates structured data that reduce subjectivity, enhance comparability, and greatly improve communication and cooperation among the different stakeholders. Importantly, this process enables implementers not only to focus on detection of translation errors, but also to use the evaluation data for error and root cause analysis and quality improvement of the production processes involved.
Given special considerations, the system can also be used to improve the global readiness of source content and the resources used during the translation process.
This website provides comprehensive information concerning the implementation of MQM for analytic translation quality evaluation. Critical discussions include:
The approach is based on two international standards:
- ISO DIS 5060:2024, Translation services—Evaluation of translation output—General guidance
- ASTM WK46396: Standard Practice for Analytic Translation Quality Evaluation