Advanced Learner Model for Error Collection: Using Japanese Honorifics as an Example

Abstract

Over-reliance on context-unaware translation tools erodes learners' ability to detect pragmatically inappropriate expressions. This study develops a learner model for advanced learners to systematically collect and classify Contextual Errors, using Japanese Honorifics (Keigo) as a case study. The model enhances metalinguistic awareness through error visualization and provides a framework adaptable to other languages.