Abstract: 在人工智能飞速发展的今天，算力、算法和数据成为了支撑和推动人工智能不断演进的三大支柱。其中，语料库作为数据资源的核心，则可称得上是人工智能得以发展的最重要的养分。以ChatGPT为代表的各种大语言模型之所以能够具备自然语言理解和生成能力，正是基于对超大规模语料库的训练和学习。因此，近些年随着人工智能技术的大爆发，语料库也日益受到人们的关注。

DOI: 10.30275/JCLE.202606_24(1).0001

Please Cite: Lin, J. Z. 林进展（2026）前言：语料库及其在华文教学与研究中的应用（Preface: Corpora and their applications in Chinese language teaching and research），《华文学刊》（Journal of Chinese Language Education），24（1），1-4。

Abstract: In language education, the teaching of composition and the cultivation of writing skills occupy a pivotal position. Implementing automated evaluation of essays not only assists teachers in their instructional work but also significantly enhances students' learning outcomes. Although automated assessment of argumentative essays in high school has garnered widespread attention, research on Chinese argumentative essays remains relatively insufficient. This is primarily due to the complexity of annotating argumentative information in Chinese. Most existing annotation schemes are overly simplistic and homogeneous, failing to comprehensively and accurately reflect the argumentative structure of essays, and also making it difficult to train precise and reliable automated evaluation systems using annotated corpora. This study focuses on argumentative essays written by middle school students and aims to establish a practical annotation framework suitable for automated evaluation systems. In terms of argument types, a detailed classification of evidence is provided. Regarding argumentative relations, 15 types of relations are categorised from both vertical and horizontal dimensions, enabling the evaluation system to identify and analyse interactions between argumentative elements. The annotated corpus provides robust data support for system development, thereby advancing the field of automated evaluation for Chinese argumentative essays.

Keywords: argumentative essay annotation guidelines, argumentative types, argumentative relations

DOI: 10.30275/JCLE.202606_24(1).0002

Please Cite: Bai, X. P., & Zhou, X. Y. 柏晓鹏、周心怡（2026）自动作文评价中的议论文标注规范的构建（Construction of annotation guidelines for argumentative essays in automated essay scoring），《华文学刊》（Journal of Chinese Language Education），24（1），5-23。

Abstract: Singapore's Chinese textbooks encapsulate the essence of Chinese education and are deeply inscribed with the historical imprints of multicultural synthesis. This study takes two sets of 1970s primary school Chinese textbooks as samples: the 1971 Chinese Language (Second Language) for English-medium schools and the 1978 Chinese Language (Mother Tongue) for Chinese-medium schools, employing corpus linguistics to systematically examine the density, distribution, and categorical features of cultural vocabulary through quantitative and qualitative analyses. Findings reveal significant differences attributable to pedagogical positioning: the 1971 textbook exhibits higher cultural vocabulary density, adopting a compensatory strategy that establishes learner cognition through explicit material cultural vocabulary and extensive local cultural references; the 1978 textbook shows lower density, employing an implicit immersion strategy that relies on students' existing cultural environment, genre expansion, and high recurrence of core vocabulary to achieve cultural transmission. Both textbooks demonstrate higher spiritual culture and lower material culture proportions, yet the Chinese-medium school textbook significantly strengthens national identity content while reducing moral education categories. In terms of localisation, the Englishmedium school textbook features broader distribution, while the Chinese-medium school textbook concentrates on selected core concepts.

Keywords: Chinese language education, Chinese language textbooks, cultural vocabulary, corpus linguistics

DOI: 10.30275/JCLE.202606_24(1).0003

Please Cite: Luo, Q. M., Jiang, J. P., & Wu, F. H. 罗庆铭、蒋静萍、吴福焕（2026）新加坡建国初期华文教材文化取向的对比分析——以话题内容为切入点（A comparative study of the cultural orientation of Chinese language textbooks in Singapore's early years of independence: Taking topic content as the starting point），《华文学刊》（Journal of Chinese Language Education），24（1），24-46。

Abstract: This study proposes a framework for identifying Chinese academic vocabulary by incorporating Contextual Similarity (CS) as a key metric and applying cluster analysis. First, 5,052 candidate words were screened from academic and general corpora of equal size (each containing 220 million characters), using three indicators - Frequency Ratio (FR), Disciplinary Coverage (DC), and Disciplinary Distribution Uniformity (DDU). These candidates were subsequently clustered into six groups via the K-means algorithm, using CS, FR, and DDU as features. These groups were further summarised into three spectrums that reflect the overlapping relationships among general, academic, and specialised vocabulary. Building on this typology, a three-tier vocabulary system was constructed, consisting of a broad set (5,052 items), a basic set (2,499 items), and a high-frequency set (785 items). This system can provide tiered vocabulary resources for different stages of academic Chinese instruction. The study demonstrates that CS is an effective indicator for extracting academic vocabulary.

Keywords: academic Chinese, academic word list, contextual similarity, cluster analysis, continuity

DOI: 10.30275/JCLE.202606_24(1).0004

Please Cite: Li, A. 李安（2026）基于语境相似度的汉语学术词汇抽取与聚类研究（A contextual similarity-based approach to extracting and clustering Chinese academic vocabulary），《华文学刊》（Journal of Chinese Language Education），24（1），47-67。

Abstract: This study explores the linguistic operating mechanisms of classical Chinese texts and their pedagogical applications. It uses systemic functional linguistics as a theoretical framework. By integrating thematic analysis, the study analyses 32 texts spanning from the Pre-Qin period to the Qing dynasty. These texts cover genres such as poetry, Ci, memoirs, discourses, essays, and memorials. The findings reveal that, at the ideational level, cultural symbols such as "moon", "wine", and "lotus" construct the field and convey social meanings through transitivity processes and the voice system. At the interpersonal level, mood and modality systems realise the author's emotional stance and communicative strategies, while natural imagery and cultural symbols concretise emotions. At the textual level, information structures promote discourse coherence and manifest genre structures such as "introduction-body-conclusion". Based on these findings, this study proposes a discourse-analysis-oriented teaching model for classical Chinese, including strategies such as semantic mapping, modular imitation, and structural transfer, and designs a literacy assessment system grounded in the three metafunctions. This provides an innovative pathway for classical Chinese instruction that combines theoretical depth with practical value.

Keywords: systemic functional linguistics, classical Chinese texts, discourse analysis, cultural symbols, pedagogical application

DOI: 10.30275/JCLE.202606_24(1).0005

Please Cite: Zhang, Y. Z. 张燕珠（2026）基于系统功能语言学的文言语篇分析：语言运作机制与教学应用研究（Discourse analysis of classical Chinese texts from a systemic functional linguistics perspective: Linguistic mechanisms and pedagogical applications），《华文学刊》（Journal of Chinese Language Education），24（1），68-90。