A Corpus-Based Analysis of Chinese Character Errors

Abstract

Based on the corpus, this research conducts a systematic statistical and analysis on the types of Chinese characters errors of Chinese learners from different countries. It is found that errors of traditional Chinese characters are the main types of learners from Philippines, Indo-European language countries and Sino-Tibetan language countries. Learners from Japan, South Korea and North Korea have the highest frequency of wrongly written characters. To know the frequency of various types of errors in learners with different native language backgrounds could provide reference for the teaching of Chinese characters as a second language.