1. Introduction

This article introduces the background and significance of the research on the vocabulary features of Chinese language training based on analogical corpora. Translation and language contact are emerging topics in corpus translation research, including language contact in translation and research on contact language variants. Chinese language training is a type of Japanese translation of written language, with characteristics that distinguish it from traditional and written languages. Previous studies have mostly been provincial and illustrative, lacking systematicity. This article constructs a Chinese language training corpus and a Japanese language corpus, analogically examining the vocabulary characteristics of Chinese language training, providing evidence for understanding the essence and laws of Chinese language training.

2. Research background and issues

Explored the language characteristics of the Chinese language training style as a variant of ancient Japanese and its differences from Japanese texts. Scholars have different opinions on the reasons for the differences between Chinese language training and Chinese language, such as gender differences and differences between written and spoken language. Tsukishima Yuki proposed a synchronic comparison between training point materials and kana literary language, while Kobayashi Fanggui studied the language characteristics of Chinese language training point materials. However, existing research has mostly focused on linguistic formal oppositions, neglecting the overall linguistic features of Chinese language training variants. As a word for word translation method, Chinese language training has similarities but also differences with translation. The evaluation criteria for instructional texts in academia have shifted from equivalence to diverse perspectives, such as translation studies and cultural translation perspectives. The emergence of corpus translation studies has provided new ideas for Chinese language training research, focusing on the linguistic features of translated texts. This article attempts to examine the language features of Chinese language training texts from the perspectives of corpus translation studies and Japanese stylistics through vocabulary parameters, in order to explore their similarities and differences with original Japanese and provide a new perspective for understanding the essence and laws of Chinese language training. The research questions include the differences in vocabulary features between Chinese and Japanese training texts, as well as the influence of the source language text on the training results.

3. Research Design and Methods

This paper introduces the design and method of studying the vocabulary features of Chinese language training based on analogical corpora. The study utilized a self built Japanese Heian period Chinese language training corpus and Japanese language corpus, including Buddhist scriptures and Chinese texts, as well as Japanese language texts, totaling 1 million words. All corpora have undergone morphological analysis and part of speech tagging. The study uses vocabulary parameters from corpus translation studies and Japanese stylistics, such as STTR and vocabulary density, to compare and analyze the language features of Chinese and Japanese training texts. Perform multiple comparisons of different text types using statistical methods such as normalization test, independent sample t-test, and point two column correlation analysis. Use WordSmith 8.0 and SPSS 26.0 software for corpus retrieval and data statistics.

4. Results and Discussion

Through independent sample t-test and point two column correlation analysis, the vocabulary density, core word coverage, header word coverage MVR、 There are significant differences in 10 parameters, including noun rate, and the use of Japanese and Chinese words. The results showed that the vocabulary density and noun rate of the training text were higher than those of the Japanese text, while the coverage of core words, header words, and MVR were lower than those of the Japanese text. This reflects that the vocabulary range of the training text is wider, the use of commonly used words is less, the tendency of action description is obvious, the modifier is weaker, and the generalization is stronger. In addition, there are more Chinese words used in the training text, and fewer Chinese words used, which increases the difficulty of the text. These characteristics are contrary to the "simplification" feature of translation and reflect the particularity of language contact in Chinese language training.

5. Conclusion

By using the method of analogical corpus translation research, significant differences in vocabulary characteristics between Chinese training texts and Japanese texts have been revealed, including information density, vocabulary scope, summarization, and the proportion of Chinese words and language words used. The Chinese language training texts show a trend of complexity, influenced by restrictive factors such as differences in bilingual systems, training rules, target language style norms, and social context. The vocabulary patterns of Buddhist scripture and Chinese book readings are highly similar, and the influence of text type on the vocabulary characteristics of the reading results is not significant. However, the source language transmission effect still exists. The study has demonstrated the applicability of corpus translation research methods, providing materials for understanding the diversity of human communication and its limiting factors, and expanding the research on "third language codes". Future research needs to explore the language features of Chinese language training more extensively from the perspectives of semantics, syntax, and other aspects.