2022 개정 중국어 교육과정의 기본어휘 고찰

강병규 ( Kang¸ Byeongkwu ),손민정 ( Shon¸ Minjung ) 한국중국언어학회 2023 중국언어연구 Vol.- No.108

The revised 2022 Second Language Curriculum for Chinese prioritizes a core set of basic vocabulary that serves as a cornerstone for instructional and pedagogical resources. This study examines the methods used to select these vocabulary items and analyzes their frequency and syntactic characteristics. Our goal is to improve the understanding and effective use of basic Chinese vocabulary in educational settings. In 2022’s curriculum update, there has been a notable reduction in vocabulary to align with the broader strategy of minimizing learning content. As a whole, the 2022 Revised Basic Chinese Vocabulary Table comprises 865 words. The vocabulary selection underwent multiple stages. Initially, words from the 2015 Basic Vocabulary Table were screened based on their corpus frequency, dictionary inclusion. Subsequently, words from the 2022 “Basic Expressions in Communication” and those recurrent in College Scholastic Ability Test were examined, coupled with expert feedback to finalize the inclusions. The 865 words in the 2022 Revised Basic Chinese Vocabulary are characterized by the following features: (1) On average, the basic vocabulary belongs to very high-frequency words. (2) Among the basic vocabulary, words that can be categorized as content words make up the absolute majority. In comparison, functional words with grammatical functions are in the minority. (3) In terms of parts of speech, the basic vocabulary is dominated by verbs, nouns, adjectives, adverbs. In addition to the vocabulary list, we have updated and enriched accompanying illustrations to aid in effective vocabulary utilization. Notably, we have relaxed constraints on the use of 800 core vocabulary items, allowing for greater lexical flexibility. Elements like compound word morphemes, which were strictly regulated in prior versions, may now be used independently. Further enhancements include clearer annotations, added canonical character explanations, rectified confusing sign systems, and the inclusion of example words to bring specificity and clarity to interpretations.

딥러닝 언어모델과 중국어 문법 ―BERT를 활용한 방향보어의 예측 모형을 중심으로

강병규 ( Kang Byeongkwu ) 한국중국어문학회 2021 中國文學 Vol.106 No.-

In this study, we investigated how accurately the BERT model can predict Chinese directional complement. In addition, we analyzed which words the BERT model uses as an important clue in the Chinese directional complement inference process. According to the results of this study, it can be seen that the BERT model shows excellent performance in inferring distributional features and grammatical relationships based on transfer learning. Results of experiments with five Chinese directional complements show that the accuracy rate of predictions is quite high. In addition, as a result of analysis using the masked language model, it was found that the BERT model appropriately uses important clues to determine Chinese directional complement in context. We believe that this study is not only meaningful in the field of NLP, but also provides insight into Chinese grammar research or language education. If this methodology is properly utilized, it will be possible to establish an application system for Chinese grammar research and education. In Neural network models, sufficient language data learning allows us to predict which language expressions are more natural to use. Proper use of these advantages will give us insight into Chinese grammatical functions. This Chinese grammar prediction system will also help Chinese learners improve their skills by showing them what expressions are grammatically correct.

Diachronic Development of the Chinese Negative 沒 from a Typological Perspective

Eunhee Paek,Byeongkwu Kang,Dongchoon Ryu,Asa Synn,Jungku Park,Yunhui Dang 한국중어중문학회 2014 中語中文學 Vol.58 No.-

Two factors contributed to the emergence of mei 沒 in the course of diachronic development of Chinese. The first factor is semantic and functional in its nature. By the mechanism of metonymy, the meaning of mei 沒 as a main predicate was extended from “to sink” and “to disappear” to “not to exist.” The second factor is phonological. Wu 無 and wei 未, which had been used as an existential negative and a perfective negative respectively, lost the [m]-initial, and consequently their status as m-negatives as well. This caused mei 沒, which semantically was undergoing grammaticalization into an existential negative, to substitute for the pre-existing existential negative wu 無 first. After the property of boundedness is firmly established in Chinese predicates later on, mei 沒 extended its meaning and function to become a perfective negative, finally attaining the status as a new m-negative. However, the substitution of existential and perfective negatives by mei 沒 mainly occurred in the Northern dialects only. In many of the Southern dialects in which the [m]-initial of wu 無 and wei 未 was retained, an existential negative and a perfective negative are still distinct, and no substitution by mei 沒 has occurred. Another noteworthy aspect of the development of the negatives in Chinese is that it reflects the development of Chinese from a synthetic to analytic language. A negative with a complex function developed into a negative with the sole function of negating, as illustrated by the development of mei 沒, meiyou 沒有, and mei 沒 V. The development of the negative pairs such as wei 未 vs. weiceng 未曾, wu 無 vs. wuyou 無有, fei 非 vs. feishi 非是 can be viewed in the same vein. The analyticization process resulted in the reduction of the multitude of negatives to just two, namely, bu 不 and mei 沒.

말뭉치 기반 중국어 문두 다중논항의 의미역 배열 경향성 고찰

박정구 ( Park Jungku ),강병규 ( Kang Byeongkwu ),박민준 ( Park Minjun ),유수경 ( Yu Sukyong ) 한국중국언어학회 2023 중국언어연구 Vol.- No.106

This paper examines the ordering patterns of multiple arguments preceding predicates in Mandarin Chinese. The focus of the study is to investigate how the order of multiple arguments at the beginning of sentences affects their acceptability, and to identify any constraints that may exist. To achieve this goal, the study extracted about 5,000 sentences from modern Chinese corpus data of various genres in which multiple arguments appear at the beginning of a sentence as topics, and conducted a statistical analysis of the word order arrangement of arguments according to different semantic roles. Through the analysis of this paper, the word order relationship between arguments with different semantic roles could be identified as shown below. First, Time/Locative precedes all semantic roles, and Time tends to precede Locative. This is a consequence of the fact that the Time/Locative functions as a theme providing a background for the information of the rest of the sentence. Second, it was confirmed that there was a certain tendency in the word order between obligatory arguments. In general, the semantic roles used as the subject before the verb in the SVO sentence are preceded in the topic sentence by the semantic roles used as the object after the verb. In the opposite case, it is accompanied by contrastive meaning and has strong restriction on acceptability. Third, the optional arguments as topic can be divided into Instument, Manner, Material, Causes, Reason, and Source. Instrument, Manner, Material and Spacial Source tend to follow obligatory arguments, and Causes, Reason, and Temporal Source tend to precede obligatory arguments. This can be explained by the restriction based on cognitive iconicity.

Towards Understanding and Applying Chinese Parsing using Cparser

PARK Minjun(朴敏浚),KANG Byeongkwu(姜柄圭) 한국중어중문학회 2020 한국중어중문학회 우수논문집 Vol.- No.-

This paper focuses on parsing processes and principles, which are essential tasks for machines to understand syntactic and semantic structures of a sentence. Machine analysis procedures for Chinese sentences, including word segmentation, and part-of-speech tagging and parsing, were visually represented using Cparser, a rule-based constituency parser developed by Peking University. Then we explained how linguistic knowledge was embodied in Cparser lexical, syntactic and semantic components, and discussed their complex interplay that allows automatic parsing. As a practical example, a Chinese textbook treebank is also constructed using Cparser. According to the theoretical and practical discussion in this paper, Peking University Cparser can easily to reflect and modify undated linguistic knowledge and is expected to be widely used as an analysis and verification tool for Chinese grammar research.

중국어 방언의 완정상·완료상·완성상 표지의 분포와 그의 언어유형론적 함의

박정구 ( Park Jungku ),강병규 ( Kang Byeongkwu ),유수경 ( Yu Sukyong ) 한국중국언어학회 2020 중국언어연구 Vol.0 No.87

In this paper, we have investigated the variations of perfective and perfect aspect makers in 135 Chinese dialects corresponding to ‘le(了)’ in Mandarin. The aim of this study is to compare the variation forms of aspect markers between various Chinese dialects and to examine the correlation between form and function. In Mandarin Chinese, it is difficult to distinguish semantic functions because perfective and perfect aspects are represented by the same ‘le(了)’. However, in Chinese dialect, the marker corresponding to ‘le(了)’ has various forms depending on the syntactic distribution and semantic function. Depending on the dialect, ‘le₁(了1)’ following the verb and ‘le₂(了2)’ of sentence-final may have one form or two or more forms. In order to observe these aspects comprehensively, we reviewed related prior studies and conducted a survey of more than 135 dialect speakers. According to a survey of Chinese dialect, dialects using one form make up 40% of the total. The remaining 60% of the dialects have two or more forms. In a survey of Chinese dialects, we found that aspect markers corresponding to “le” vary in distribution and combine with different types of predicates. And we can see that the distributional difference of ‘le(了)’ greatly affects the functional difference. Specifically, the function of ‘le(了)’ has a clear contrast between ‘sentence-final’ and ‘non-sentence-final’. In other words, functional contrasts are basically closely related to the distribution of ‘le(了)’. When ‘le(了)’ is placed in sentence-final, it is closely related to the form of ‘le₂(了2)’, whether it is added to a verb or an adjective. On the other hand, when ‘le(了)’ is used in a non-sentence-final position, it is closely related to the form of ‘le₁(了1)’, whether it is added to a verb or an adjective. In addition, the contrast between the meanings of predicates expressed as ‘dynamic’ and ‘static’ is not as obvious as the contrast between ‘sentence-final’ and ‘non-sentence-final’. This paper classifies types of 135 dialects based on the formal similarity of ‘le(了)’ As a result of statistical cluster analysis, it can be seen that it is divided into three types. These three types correspond to perfective, perfect, and completive, respectively. (1) ‘le₁(了1)’ is an aspect marker that can be attributed to perfective. As a result of the dialect investigation, the sentences corresponding to ‘le₁(了1)’ can be classified into the same group because they maintain similarity in form. (2) ‘le₂(了2)’ is an aspect marker that can be classified as perfect. According to the dialectal survey, ‘le₂(了2)’ was classified as an independent type, although its internal variation varied but had a distinctive morphological characteristic from ‘le₁(了1)’ as a whole. (3) ‘le(了)’ which has the function of result-complement belongs to the completive aspect because the degree of grammaticalization is lower than the other two types. When ‘le(了)’ is combined with a verb that has meaning of destruction and consumption, it can be interpreted as completive aspectual meaning. The degree of grammaticalization of these three types can be said to have a hierarchy of ‘completive <perfect <perfective’ as discussed in previous studies. In addition, ‘le(了)’, which has an aspectual meaning of completive/perfect/perfective, implies an aspectual meaning of inchoative in the interpretation of the bounded event structure. In this paper, we have classified and analyzed the various aspectual types and systems of ‘le(了)’ in Chinese dialects. In the future, we will try to compare and study perfective/perfect in other languages. This study on the perfective, perfect, completive aspect with empirical evidence and theoretical explanation shows that the development of the Chinese aspectual system has been following the universal tendency of human language evolution along the path of grammaticalization of aspect markers.

어순 수용성 데이터 기반의 중국어 문두 다중논항 의미역 배열 기제 및 위계 연구

박정구 ( Park Jungku ),강병규 ( Kang Byeongkwu ),박민준 ( Park Minjun ),유수경 ( Yu Sukyong ) 한국중국언어학회 2023 중국언어연구 Vol.- No.109

This paper extracted a large number of sentences with multiple arguments at the sentence beginning position from various genres of modern Chinese corpus data through coding programs as well as artificial methods. After changing the order of arguments of sentence-initial position in the extracted sentences, word order acceptability data was constructed through verification by native speakers. For the large amount of data constructed in this way, the quantity and ratio of combinations of semantic-roles that the multiple arguments at sentence-initial position play were statistically analyzed. Based on the analysis so far, it has been possible to induce the constraints on the multi-argument arrangement and establish the hierarchy of the semantic role for topicalization. Among obligatory arguments that appear at the beginning of a sentence, the arguments used as the object precede the arguments used as the subject. Among optional arguments, Time, Locative, and Cause arguments precede other arguments. Source and Goal arguments have something in common: they tend to precede other arguments when they express time. However, when they represent locative meaning, they shows different characteristics. The locative Source follows the obligatory arguments, and the locative Goal precedes the obligatory arguments. Instrument, Manner and Material are closest to the predicate, and Beneficiary and Recipient precede obligatory arguments. The word order of multiple arguments at sentence-initial position is influenced by iconicity, semantic relationship with the verb, and whether or not they are obligatory arguments.

신경망 언어 모델 구축을 통한 중국어 문두 다중논항 어순 수용성 예측 연구

박정구 ( Park Jungku ),유수경 ( Yu Sukyong ),박민준 ( Park Minjun ),강병규 ( Kang Byeongkwu ) 한국중국어문학회 2024 中國文學 Vol.118 No.-

본고에서는 중국어 문두 다중논항의 어순 배열 문제를 해결하기 위해 트랜스포머 기반의 신경망 언어 모델을 적용하고 그 효율성을 검증하는 연구를 수행하였다. 본 연구에서는 문두에 나타나는 다양한 성분들과 이들 사이의 상호 작용, 그리고 어순 배열에 대한 제약이 어떻게 문법성을 형성하는지에 초점을 맞췄다. 이를 통해 중국어 문두 다중논항의 어순 배열과 관련된 문법성을 판단하는데 자연어처리 방법론이 얼마나 효과적인지 검증하고자 하였다. 본 연구에서 구축된 ‘문두 다중논항 데이터셋’은 15,298개의 중국어 문장으로 이루어져 있으며, 이 데이터셋은 문법적으로 올바른 문장과 비문법적인 문장을 이진 분류 방식으로 레이블링 하여 사용되었다. 연구팀은 이 데이터셋을 활용하여 문두 다중논항의 어순 수용성을 예측하는 모델을 개발하고, 그 정확도를 평가하였다. 연구 결과, 모델은 90%의 높은 정확도로 어순 수용성을 예측할 수 있음을 보여주었다. 또한, 본 연구는 원어민의 판단과 모델의 예측이 일치하지 않는 경우를 분석하여, 언어 수용도가 단순한 이분법으로만 구분되지 않음을 확인하였다. 이를 통해 언어 모델의 예측 능력과 원어민 판단 사이의 상관성을 보다 정교하게 파악할 수 있었다. 원어민의 판단과 AI 모델의 예측이 일치하지 않는 경우의 다양성을 고려하여, 본 연구는 153개의 문장에 대한 2차 분석을 실시하였다. 이 과정에서 74개의 문장이 유사 문장으로 재분류되었고, 결과적으로 1,450개 문장(94.8%)이 두 예측치가 일치하거나 유사한 것으로 평가되었다. 이는 AI 모델 예측의 높은 정확도를 보여주는 결과이다. In this paper, we applied a transformer-based neural network language model to solve the word order problem in Chinese Sentence-Initial Multiple Arguments (SIMA) and verified its effectiveness. The study focused on the different constituents that occur at the beginning of Chinese sentences, their interactions, and how word order constraints contribute to grammaticality. The goal was to validate the effectiveness of natural language processing methods in determining the grammaticality of word order in Chinese SIMA. To investigate this, we constructed a specialized dataset and developed a predictive model. The SIMA dataset consists of 15,298 Chinese sentences. These sentences were labeled as grammatically correct or incorrect using a binary classification method. We developed a model to predict the acceptability of word order in SIMA and evaluated its accuracy. The model demonstrated a high level of accuracy in predicting word order, achieving 90% correctness compared to established grammatical norms. In addition, the study analyzed cases where the model’s predictions differed from native speakers’ judgments, showing that acceptability prediction is not just a simple binary distinction process. This led to a more nuanced understanding of the relationship between language model predictions and native speaker judgments. In the secondary analysis focusing on sentences where the AI model’s predictions diverged from native speaker judgments, we found that after reevaluation, a significant proportion of these sentences (94.8%) aligned closely with native judgments, highlighting the nuanced capabilities of our model. These findings not only demonstrate the effectiveness of our model in analyzing Chinese SIMA but also contribute to a deeper understanding of AI’s role in linguistic analysis.