Abstract
The aim of this study is to develop an error-tagging support system for the construction of a Japanese language learner corpus. This study implements a proposed system ([[cite to previous presentation ・ ie., "Hashimoto, 2005"]]) and conducts an evaluation experiment. The system identifies errors by means of a hierarchically structured set of error items and a database of errors. The database was created by analyzing errors for declinable words made by Japanese language learners. The evaluation experiment investigated whether the system is capable of selecting correct items for identified errors.
