Skip to content

    METEOR

    AI fundamentals & modelsMetric for Evaluation of Translation with Explicit ORdering
    Reading time: 2 min.Editorially reviewed on:
    ← All terms

    Definition

    METEOR compares machine translations with reference texts.

    The metric combines word matches including stems or synonyms and considers word order. It provides an automatic comparison signal.

    Simply explained

    It evaluates not only identical words but also related expressions and their sequence.

    Why it matters in practice

    This allows translation systems to be compared complementarily to human evaluation.

    Example from business

    Hypothetical example: A team evaluates two translation versions based on the same references.

    Limits and pitfalls

    • Values depend on language, resources, and implementation. Good free formulations can be undervalued.

    Origin & history

    Banerjee and Lavie introduced METEOR in 2005.

    Use in mid-sized companies

    • SMEs can monitor translation workflows with sampling and automatic metrics.

    Frequently asked questions

    Does METEOR replace professional translation review?

    No, terminology and meaning require separate review.

    Sources

    Editorially reviewed on:

    On this page

    Next step

    Put knowledge into practice