• Industrial Engineering and Management Systems
• /
• 제3권1호
• /
• pp.85-91
• /
• 2004

The writing styles of TIME and Newsweek are analyzed using a specially developed linguistic program. These two news magazines were chosen because of their wide popularity. As for the results, it became obvious that both the frequency curve of words and that of characters have not changed for the past 60 years. Also, we have found that the frequency curves have some inflection points and that the genre of English writings can be identified by these points. After counting the percentage of required vocabulary for junior high school students and high school students in English writings, we can derive the relative difficulties of them using fuzzy reasoning. Fuzzy rules are constructed using features of the characteristic curves. We feel it would be a good guide index when selecting textbooks or supplementary readers.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2003년도 정기총회 및 학술대회
• /
• pp.3-6
• /
• 2003

방송 서비스상의 교육용 어학 컨텐츠의 증가와 더불어 비디오 컨텐츠의 효율적인 제공, 이용 및 관리를 위한 내용 기반 요약에 대한 연구가 필요하다. 본 논문에서는 교육용 어학 비디오의 내용 기반 요약을 위한 방법을 제안한다. 디지털 비디오로부터 샷 경계를 추출한 후 각 샷을 대표하는 키프레임으로부터 MPEG-7 비주얼 특징 값들을 추출한다. 추출된 특징 값들의 다중 조합을 통해 교육용 어학 비디오의 내용 정보를 세분화하여 요약 결과를 생성한다. 외국어 회화 컨텐츠에 대해 실험하여 알고리즘의 효용성을 검증하였으며. 제안한 방법은 교육용 방송 컨텐츠의 다양한 서비스 제공 및 관리론 위한 비디오 요약 시스템에 효율적으로 이용될 것이다.

• 한국어학
• /
• 제80권
• /
• pp.125-150
• /
• 2018

This paper discusses several prerequisites for recording other languages by using Hunminjeongum (Hangeul). The scientific, systematic, and practical features of Hunminjeongum is undoubtedly obvious. Such a view of Hunminjeongeum is based on its iconicity, featural property and non-linearity. However, the existing conviction that Hunminjeongeum can transcribe all human sounds or that Hunminjeongeum is the only script which is easy for anyone to learn requires a more rigorous consideration. Given the fact that all scripts can be evaluated as equal, without one being more superior than the other, all scripts are optimized for recording the language in its own given linguistic context. Further, the function of a script is not just limited in a simple representation of a language. Adding new letters into the system of Hunminjeongeum or maintaining its current mode of notation can be problematic. First, it is important to solve computer-related issues: the registration of Hunminjeongum into the computer system as well as its output, problems regarding unicode, the new developments required for input devices, keyboards, fonts, and computer programs.

• Journal of Information Processing Systems
• /
• 제15권5호
• /
• pp.1171-1178
• /
• 2019

This study proposes a deep neural network model based on an encoder-decoder structure for visual dialogs. Ongoing linguistic understanding of the dialog history and context is important to generate correct answers to questions in visual dialogs followed by questions and answers regarding images. Nevertheless, in many cases, a visual understanding that can identify scenes or object attributes contained in images is beneficial. Hence, in the proposed model, by employing a separate person detector and an attribute recognizer in addition to visual features extracted from the entire input image at the encoding stage using a convolutional neural network, we emphasize attributes, such as gender, age, and dress concept of the people in the corresponding image and use them to generate answers. The results of the experiments conducted using VisDial v0.9, a large benchmark dataset, confirmed that the proposed model performed well.

• Asia pacific journal of information systems
• /
• 제31권2호
• /
• pp.197-219
• /
• 2021

Social media platforms have become prominent channels for e-commerce, and the role of social network sites' (SNS) content marketing is expanding as a strategic marketing communication approach to attract and retain consumers and increase sales. In this study, we focused on South Korea market and explored the influence of linguistic complexity and informality on consumer engagement. In particular, we identified the importance of complexity, focusing on its negative effects, as well as the moderating effect of commerce features to minimize these effects. Specifically, content length, hashtags, long words, and average sentence length significantly and negatively impacted consumer engagement. The influence of emojis, an informality variable, was not statistically significant. Shoppable tags, a commerce feature that provides both advertising explicitness and shopping convenience, were a moderating factor in the influence of complexity. Our findings provide new insights for content marketing researchers, and have practical implications for social media managers and content developers.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제36권5호
• /
• pp.420-429
• /
• 2009

특허 문서는 과학기술 발전을 탐지하고 기존 트렌드를 이해함으로써 미래의 트렌드를 예측하는데 유용한 자원이다. 본 연구에서는 단위 기술을 "문제점"과 "해결방법"으로 구성되어 있다고 보고, 언어적 단서(linguistic clue)와 언어 모델(1anguage model)을 결합한 혼합 모델을 사용하여 이들에 해당하는 의미 핵심문구(semantic keyphrase)를 찾고, 의미 핵심문구로 표현되는 단위 기술을 추출하였다. 추출된 결과에 근거하여 비지도 학습(unsupervised learning) 방법으로 과학기술들의 트렌드를 발견하는 새로운 접근방법(Technological Trend Discovery, TTD)을 제안한다. 실험 결과에 따르면 본 연구에서 제안한 방법으로 과학 기술을 나타내는 의미적 핵심 문구를 추출하는데 77%의 R-정확률을 달성하였고 결과적으로 의미있는 과학기술 트렌드를 발견할 수 있었다.

• Acta Via Serica
• /
• 제2권1호
• /
• pp.31-47
• /
• 2017

The corridors to the north and south of the $Takl{\bar{a}}m{\bar{a}}k{\bar{a}}n$ (塔克拉瑪干 Ta-ke-la-ma-gan) Desert are the most important regions for cultural confluence on the Silk Road, where caravans made it to the Chinese capital or the Korean Peninsula by the northern road, through the city of Turfan, or the southern path of Khutan. Being an important part of the Silk Road in the course of history, this region was heavily influenced by the cultures of various nations and ethnic communities whose merchants utilized the road to advance their business. The region's language, writing system and literary structure were also affected, so much so that in the course of its tumultuous history, many words, phrases and terms belonging to neighboring cultures found their way into the region, leaving their mark on its linguistic structure. Of the cultural exchanges that took place between the peoples of the region, conspicuous traces can be seen in the architecture, music, literature, texts, and inscriptions. Located in the Turfan region, the minaret of Su Gong (蘇公 Su Gong ) is host to an inscription which bears many signs of such exchanges. As so far no independent research has been conducted to identify the cultural, literary and structural features conveyed in this inscription, the present paper is an attempt to study the inscription in terms of the script, language and syntax in order to unravel the effects of cultures prevalent on the Silk Road on this particular inscription. This study mainly aims to investigate the linguistic structure of the inscription and the impact of the Persian language on Silk Road culture. In fact, we approach the inscription as a symbol of cultural exchange on the Silk Road and will focus on the tradition of Persian inscription-making which affected the Turfan inscription.

• International Journal of Computer Science & Network Security
• /
• 제22권11호
• /
• pp.157-162
• /
• 2022

Cerebral palsy is one of the most serious forms of disorders of the psychophysical development of children, which manifests itself in disturbances of motor functions, which are often combined with speech disorders, other complications of the formation of higher mental functions, and often with a decrease in intelligence. The article will discuss the speech disorder in children with cerebral palsy. Emphasis is placed on some important aspects, which should bear in mind, investigating the problem of specifics of speech development of children with cerebral palsy. In particular at the heart of speech disorders in the cerebral palsy is not only damage to certain structures of the brain, but also the later formation or underdevelopment of those parts of the cerebral cortex, which are of major importance in linguistic and mental activity. This is an ontogenetically young region of the cerebral cortex, which is most rapidly developing after birth (premotor, frontal, temmono-temporal). It is important to take into account, that children with cerebral palsy have disturbances of phonemic perception. Often, children do not distinguish between hearing sounds, cannot repeat component rows, allocate sounds in words. At dysarthria, there are violations of pronunciation of vowel and consonant sounds, tempo of speech, modulation of voice, breathing, phonation, as well as asynchronous breathing, alignment and articulation. As a result, we identified the main features and specifics of the speech development of children with cerebral palsy and described the conditions necessary for the full development of language. Language disturbances in children's cerebral palsy depend on the localization and severity of brain damage. Great importance in the mechanism of speech disorders has a pathology that limits the ability of movement and knowledge of the world.

• 예술인문사회 융합 멀티미디어 논문지
• /
• 제5권2호
• /
• pp.157-165
• /
• 2015

본 연구는 다문화가정 아동의 문식성을 알아보고 문식성에 영향을 주는 인지-언어적 변인이 무엇인지 밝히고자 하는 목적이 있다. 연구대상은 전라남도에 거주하는 초등 저학년 46명, 초등 고학년 39명, 전체 85명이다. 다문화가정 어머니의 한국 거주기간은 4~16년이고, 국적은 필리핀, 일본, 중국, 베트남, 캄보디아, 태국, 우즈베키스탄, 스리랑카이다. 본 연구의 결과는 다음과 같다. 첫째, 다문화가정 아동의 읽기, 쓰기, 인지-언어적 변인들은 학년이 높을수록 수행능력이 높았다. 둘째, 다문화가정 아동의 읽기 예언변인이 시지각 하위변인인 시각형태항상성으로 나타난 것은 읽기의 선행조건에서 세부특징이 중요하다는 것을 의미한다. 따라서 다문화가정 아동의 읽기는 친숙한 단어보다는 세부특징에 대한 단서를 잘 아는 것이 중요하다고 할 수 있다. 한글을 처음 배우는 단계에서 통글자 위주의 지도보다는 자음과 모음의 특징을 인식시키는 훈련이 먼저 필요함을 시사한다.

• 한국노년학
• /
• 제31권3호
• /
• pp.751-765
• /
• 2011

본 연구는 노년의 호명과 관련해 한국사회의 다양한 언어적 실천이 노정하고 있는 갈등과 그 정치적 함의를 밝히고, 이러한 현상이 벌어지고 있는 원인을 문화적 차원에서 조명하려 한다. 근래에 들어 한국사회에서 노년을 어떻게 호명할 것인지가 일상에서 문제시되고 있다. 이에 따라 새로운 호명방식이 등장하고 있으며, 이러한 현상은 노년의 주변화를 의미한다. 하지만 새로운 호명방식들도 노년 호명의 문제를 풀어내지 못하고 있다. 그 이유는 한국사회의 연령주의적, 성차별적, 신분주의적 시선이 노년에 대한 다양한 호칭과 지칭의 의미를 주변화시키고 있기 때문이다. 또한 노년의 주변화는 문화적 차원에서 노년의 타자화가 심화되면서, 노년이 타자성의 집적지가 되어가는 '주변성의 노년화'가 진행되고 있기 때문이다. 학계와 전문가 집단에서도 노년의 호명과 관련된 다양한 언어적 실천이 지니는 정치적 함의에 대한 이해와 더불어 차별적 언어사용을 넘어서려는 실천이 요구된다.