http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
박민준 ( Park Minjun ) 한국중국언어학회 2022 중국언어연구 Vol.- No.102
This paper introduces and explains in detail the overall information and tutorial about the commonly used Chinese morphological analyzers (e.g. ICTCLAS, Jieba, Stanford CoreNLP) which are employed in Chinese preprocessing tasks of Chinese Word Segmentation (CWS) and Part-of-speech tagging. In particular, the usability of the tools was enhanced by developing simple executables distributed to linguistic researchers unfamiliar with coding, along with rich execution examples in GUI and CLI environments. Plus, by introducing the unique features and functions of each morphological analyzer, it was recommended the most suitable analyzer tailored to the needs of individual researchers. As a guide for Chinese morphological analysis, which is inevitably accompanied by data-driven quantitative research, this study presents practical tools and useful guidelines for Chinese text preprocessing to researchers who want to expand their research interests to corpus linguistics, computational linguistics, and natural language processing.
웨이보에 투영된 역정(疫情) ― COVID-19 관련 키워드의 SNS 빅데이터 분석
박민준 ( Park Minjun ) 한국중국언어학회 2020 중국언어연구 Vol.0 No.90
This paper investigates some of dominant linguistic and social trends of China in the context of COVID-19, through a comprehensive analysis of vocabulary usage patterns on Weibo, the biggest SNS platform in China. First, we introduce generative and informatic features of Weibo to highlight its distinctive characteristics from other media. Next, based on the actual utterances of 516 million Weibo users, we examined the individual linguistic expressions and collective mentality of Chinese people under COVID-19. In detail, we established COVID-19 Weibo big data which consists of 136,585 Weibos of 22,985,364 characters, and gave a qualitative and quantitative analysis from four different functional perspectives - information seeking, emotional release, citizenship behavior and social connection. In this analysis, we clearly demonstrated some widespread linguistic and cultural phenomena, such as the spread process of Chinese naming for COVID-19 and ‘GuanZhuangBingDu’ being widely used, broad textual tendency to avoid negative emotional expressions and prefer conceptual metaphors, and common types of social connections through hashtags.
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.
박민준(Park. Minjun),진호상(Jin. Hoshang),이건희(Lee. Gunhee),황광규(Hwang. Kwangkyu),김우섭(Kim. Woosup),이재호(Lee. Jaeho) 전력전자학회 2015 전력전자학술대회 논문집 Vol.2015 No.07
This paper is about the suggestion for the development in the commercialization for 3.6kW Class On-Board charger. It is suggesting non-insulation AC-DC Boost Power Factor correction circuit and insulation DC-DC resonant Converter for circuit design. In addition, Input AC voltage in the power supply is DCM control which can be designed to decrease the inductance for the inductor size to be reduced. DCM controls and Interleaved PFC can be designed to decrease the inductor size increasing the power conversions. Also, using the insulation DC-DC resonant converter, the efficiency can be increased. This system is verified using prototype hardware.
텍스트 의미 분석 시스템의 구현과 활용 -UCREL Semantic Analysis System을 중심으로-
박민준 ( Park Minjun ) 덕성여자대학교 인문과학연구소 2024 인문과학연구 Vol.38 No.0
비록 최근 딥러닝 기술의 발전이 대규모 언어 모델[LLM] 기반 자연어 이해[NLU]라는 돌파구를 마련해주었지만, 여전히 특정한 문맥과 맥락 속에서 언어모델은 과거와 비슷한 실수를 반복한다. 이를 보조하기 위한 수단으로, 개념 체계 기반 의미 분석 시스템 [Concept-Based Abstracting system]은 자연어처리에 필요한 개념 정보를 제공함으로써 복잡다단한 자연어의 자동 의미 분석을 가능하게 한다. 본 연구는 개념 체계 기반 의미 분석 시스템의 하나인 랭캐스터 USAS[UCREL Semantic Annotation System]의 내부 구조와 작동 원리 및 활용 예시를 자세히 살펴본다. USAS는 본문에서 제시한 의미 중의성 해소[WSD], 텍스트 분류, 키워드 분석 등을 통해 자동화된 내용 분석[automatic content analysis]을 가능하게 하며, 이는 단순 반복 작업만을 수행하는 기계적인 시스템이 아니라 한층 발전된 범용적[versatile] 언어 인공지능으로 나아가는 계기를 마련해 준다. Although the recent advancement in deep learning technology has provided a breakthrough in LLM-based natural language understanding, language models still repeats mistakes similar to those of past models in certain contexts. As a means to guide them, the Concept-Based Abstracting system enables automatic semantic analysis of complex natural language by providing conceptual information necessary for natural language processing. This study examines in detail the internal structure, operational principle, and practical examples of Lancaster USAS (UCREL Semantic Annotation System), one of the concept-based semantic analysis systems. USAS enables automated content analysis through tasks such as disambiguation of meanings presented in the text (WSD), text classification, keyword analysis, etc. This goes beyond a mechanically repetitive system, serving as a significant step towards a more advanced and versatile language artificial intelligence.