introduction

Emphasis was placed on the importance of vocabulary in second language learning, distinguishing between receptive and productive vocabulary, and highlighting the significance of productive vocabulary research in terms of complexity, accuracy, and fluency in oral and written expression. At the same time, it pointed out the shortcomings of non universal languages such as French in the study of vocabulary complexity, especially in the research of written language. The aim of this study is to investigate the correlation and developmental characteristics between vocabulary complexity and French proficiency of learners at different levels through a longitudinal study of the compositions of French major students in a domestic university, providing a new perspective for French teaching and research.

1、 Literature Review

Discussed the assessment of language complexity in second language acquisition research, with a particular focus on vocabulary complexity. Vocabulary complexity is quantified through three indicators: vocabulary density, diversity, and complexity, which respectively reflect the amount of text information, the number of different word types, and the proportion of infrequently used vocabulary. Vocabulary density was proposed by Ur, with written texts typically exceeding 40% and spoken texts below 40%. Vocabulary diversity was evaluated through methods such as TTR, HD-D, and MTLD. HD-D showed a high correlation with French proficiency test results, while MTLD was sensitive to text length. The complexity of vocabulary is reflected through the frequency profile, which reflects the learners' written vocabulary. Research on the complexity of second language vocabulary both domestically and internationally includes measurement validity, observational studies, and experimental studies. McCarthy and Jarvis validated the effectiveness of vocabulary diversity measurement methods, while Bao Gui introduced advanced vocabulary diversity measurement methods. Bei Xiaoyue explored the impact of topic familiarity, task preparation, and English proficiency on the complexity of second language vocabulary. At present, research mainly focuses on second language English, with less research on non universal languages such as French, especially written language. Research on the complexity of French vocabulary internationally has focused on spoken language, with less attention paid to written language. There is relatively little research on the vocabulary complexity of French learners in China. Qin Jie and Wu Wangjiao explored the interaction effect between topic familiarity and learner creativity. Longitudinal studies can provide important new insights into the development of learners' vocabulary abilities over time, but currently little is known about the changes in vocabulary complexity of French learners over time. This study focuses on the complexity of French vocabulary and uses natural language processing techniques to measure the vocabulary density, diversity, and complexity of learners at two learning stages. It explores the correlation between changes in different measurement indicators and French proficiency, and examines the development of learners' vocabulary abilities at different learning stages.

2、 Research Methods

This chapter introduces the methods for studying the development of vocabulary complexity in French learners' compositions. The study adopts a mixed design of 3 (French proficiency) × 2 (learning stage) to explore the changes in vocabulary complexity of students at different levels during different periods. The participants are 68 third year French major students from a university in East China, whose native language is Chinese. The research corpus comes from the classroom compositions of participants in the fifth semester, with a total of 136 valid compositions collected. This study follows Lu Xiaofei's handling of spelling errors and uses TreeTagger for part of speech tagging and word form restoration. Referring to Bult é and Housen's classification of vocabulary complexity, we comprehensively consider three measurement indicators: vocabulary density, diversity, and complexity. Vocabulary density is measured as the ratio of the number of actual word symbols to the total number of word symbols in a text. Vocabulary diversity is measured using the HD-D value, and vocabulary complexity is measured as the ratio of word class symbols other than commonly used high-frequency words to the total number of word symbols. Statistical analysis was conducted using a multivariate mixed design and repeated measures multivariate analysis of variance was performed using R software.

3、 Result

The development of vocabulary complexity in French learners' compositions was explored through multivariate statistical analysis. Research has found that French proficiency and learning stage have a significant impact on vocabulary complexity, but there is no interactive effect between the two. The vocabulary complexity of the high-level group is significantly higher than that of the low-level group, while vocabulary diversity and complexity are the main factors causing the differences, and the influence of vocabulary density is not significant. The change in vocabulary density is not significant at different learning stages, while vocabulary diversity and complexity increase with the improvement of French proficiency, and T2 stage is significantly higher than T1 stage. The learning stage has no significant impact on vocabulary diversity, but French proficiency has a significant effect. The complexity of vocabulary is significantly influenced by French proficiency and learning stage, with the high-level group showing significantly higher vocabulary complexity in T2 stage than in T1 stage.

4、 Discussion

This chapter discusses the impact of French proficiency and learning stage on vocabulary complexity and its measurement indicators. Research has found that there are significant differences in vocabulary diversity and complexity among different French proficiency groups, but there is no significant change in vocabulary density. Vocabulary density cannot effectively distinguish the language proficiency and developmental stages of learners, while vocabulary diversity and complexity increase with language proficiency. The coverage of high-frequency words in learners' compositions is high, and the complexity of vocabulary increases with language proficiency, but it fails to effectively distinguish between low and medium level groups. The selection of a word frequency table is crucial for measuring vocabulary complexity, taking into account the learner's native language background and language input environment. The phased development of learners is influenced by their initial language proficiency, and progress is relatively faster in the short term. Teachers should integrate flexible use of vocabulary into daily teaching, especially in oral and writing activities. The measurement method of English vocabulary complexity has reference value in French, but it needs to be flexibly adjusted and optimized according to the learner's background and characteristics.

conclusion

This study compared and analyzed the compositions of third year French major students in a university during one semester, revealing the developmental characteristics of learners' vocabulary complexity in their compositions. The results showed that language proficiency had a significant impact on vocabulary complexity, with the high-level group outperforming the low-level group in terms of vocabulary diversity and complexity. The impact of the learning stage on vocabulary complexity is mainly reflected in the increase of complexity. The research results have reference value for French vocabulary teaching. Teachers can use indicators of vocabulary complexity and diversity to evaluate students' vocabulary development and flexibly adjust teaching strategies. Students' vocabulary complexity has significantly improved in the short term, but diversity still needs to be improved, and more language input should be sought. There are shortcomings in the research, such as short period, limited sample size, and restricted data sources. Future research should extend the period, expand the sample size, and combine complex dynamic system theory to conduct more comprehensive research from multiple perspectives such as syntax, morphology, and collocation construction.