Shortening in Korean

Weon-Don Jeong 한국음운론학회 1997 음성·음운·형태론 연구 Vol.3 No.-

Jeong, Weon-Don. 1997. Shortening in Korean. Studies in Phonetics, Phonology and Morphology 3. 265-277. This paper deals with shortening phenomena in Korean. Among them, we examine formation of shortening, types of shortening, and properties of shortening. First, we show that shortening is often formed by partial deletion and attachment, and sometimes by glide formation. Second, we discuss that shortened forms are found not only in nouns and verbs but also in words and phrases, and that they occur in an optional way and in an obligatory way. Finally, we investigate that shortening is shown in casual style, that it changes form, that it is regarded as a process of deletion and attachment, and that it is not governed by phonological condition. (Semyung University)

A Study on the Hopfield Neural Scheme for Data Association in Multi­Target Tracking

Lee, Yang­-Weon The Korea Institute of Information and Commucation 2003 한국정보통신학회논문지 Vol.7 No.8

In this paper, we have developed the MHDA scheme for data association. This scheme is important in providing a computationally feasible alternative to complete enumeration of JPDA which is intractable. We have proved that given an artificial measurement and track's configuration, MHDA scheme converges to a proper plot in a finite number of iterations. Also, a proper plot which is not the global solution can be corrected by re­initializing one or more times. In this light, even if the performance is enhanced by using the MHDA, we also note that the difficulty in tuning the parameters of the MHDA is critical aspect of this scheme. The difficulty cat however, be overcome by developing suitable automatic instruments that will iteratively verify convergence as the network parameters vary. 본 논문에서는 다중표적 추적을 위한 데이터 결합 기법 중에서 MHDA 스킴을 제안하였다. 이 구조는 기본의 JPDA보다 계산면에서 단축이 가능하여 실제 응용에 많은 적용이 기대된다. 인위적인 측정값과 표적을 이용하여 시뮬레이션을 수행한 결과 MHDA는 기존의 JPDA보다 성능도 비슷한 특성을 보이는 것을 확인하였다.

백석 시집『사슴』의 시어 양상 연구

박순원(Park, Soon-Weon) 한국시학회 2015 한국시학연구 Vol.- No.44

이 연구는『한국 현대시어 빈도사전』을 활용하여 백석 시집『사슴』의 시어의 양상을 밝히는 것을 목적으로 진행되었다.『사슴』시어의 총량, 품사별 비율, 출현빈도, 반복지수 등을 현대시 코퍼스와 대조하여 그 양상을 구체적으로 드러내는 것이다.『한국 현대시어 빈도사전』은 한국현대시 데이터베이스 구축 사업을 통해 만들어진 한국현대시 코퍼스(Korean Moderen Poetry Corpus : 약칭 KoPoCo)의 시어 빈도에 관한 여러 가지 의미 있는 정보를 모은 것이다. 『사슴』시어의 양상 중 가장 두드러진 특징은 일반명사가 많은 비중을 차지하고 있다는 점과 반복지수가 현저히 낮다는 점을 들 수 있다. 특히 반복지수는 1.58로 매우 낮은 값을 보이고 있으며, 이는 하나의 어휘에 대한 출현빈도가 2회를 넘지 않는다는 것이고, 사슴이 그 시어 총량에 비해 매우 다채로운 언어로 구성되어 있음을 보이는 것이다. 1회 출현빈도 어휘를 조사한 결과 어휘 종수 대비 75%, 출현빈도 대비 45%의 높은 비중을 확인할 수 있었다. 또한 사슴에만 쓰인 시어, 즉 백석의 개인 시어라 할 수 있는 어휘도 적지 않은데, 그 중 대부분은 일반명사가 차지하고 있다는 것을 알 수 있었다. 그리고 그 일반명사 목록을 통해 백석이 그의 초기시에 해당되는 사슴만으로도 우리 현대시어의 확장에 양적으로나 질적으로나 상당한 기여를 하고 있는 것을 쉽게 확인할 수 있었다. 『사슴』의 시어를『한국 현대시어 빈도사전』에서 대조하는 과정에서 몇 가지 문제점이 발견되었다. 단순 오타의 문제, 의미 해석의 문제 등과 더불어 누락되어 찾을 수 없는 경우가 있었고, 그 외에도 표준국어대사전에 등재된 표제어를 기계적으로 적용하여 문제가 될 만한 부분을 확인할 수 있었다. 미처 발견되지 못한 문제가 더 있을 수도 있겠으나, 현재 발견된 어휘 수로는 통계 값 전반을 무의미하게 만들 정도는 아니라고 판단된다. 그러나 이는 반드시 수정, 보완되어야 할 사항으로 보인다. The purpose of this study is to reveal the appearance of poetic words in Deer by using The frequency dictionary for the words of Korea"s modern poetry. The main content is the total amount, parts of speech rate and frequency of appearance of Deer"s poetic words, through contrast with Korean modern poetry poetic words corpus. The frequency dictionary for the words of Korea"s modern poetry is a collection of various information, which is important for the frequency of Korean modern poetry poetic words. The most important features of poetic words in Deer is that the common noun occupies a large proportion and that the number of the words are repeated significantly lower. The number of the words are repeated is 1.58. This is evidence that Deer consists of a wide variety of poetic words. Researching poetic words appeared once, it was confirmed that they are taking a very high proportion. Poetic words used only in Deer is also very much. And most of them are occupied common noun. Through the course of the study progress, it revealed several problems of The frequency dictionary for the words of Korea"s modern poetry. The problem is just misprints and words missing ect. Although minor problems, The wrong part of The frequency dictionary for the words of Korea"s modern poetry must be corrected.

사람唾液의 Cl 血型分泌型에 關한 硏究

朴淳元,韓元東 大韓法醫學會 1979 대한법의학회지 Vol.3 No.1

Moon and Wiener had demonstrated new blood factor "Cl" with anti-Cl lectin prepared from extracts of seeds of Korean Clerodendron trichotomum THUNBERG by absorption of selected human red blood cells. After that, Moon et al. had reported that Cl type specific substance in distributed in human saliva, too. Therefore, human saliva can be classified into two type, secretor of Cl (Sc) and non-secretor of Cl (nSc). To study on the frequency of Sc and nSc, relation with ABH type secretor and hereditary pattern of Cl secretors, agglutination inhibition test of anti-Cl lectin was carried out with human saliva by standard method, and the results can be summarized as follows; 1. Cl positive type can be classified secretor and non-secretor of Cl by human saliva, but Cl negative type is reacted as only non-secretor of Cl. 2. The secretor status of Cl is quite different from ABH secretor. 3. It appears that the inheritance of the saliva type of Cl follows the Mendelian law with the non-secretor of Cl transmitted through recessive genes.

벡터 양자화를 이용한 문장 및 화자 독립 감정 인식

김원구 국립7개대학공동논문집간행위원회 2003 공업기술연구 Vol.3 No.-

In this paper, the vector quantization based emotion recognition system is proposed for speaker and context independent emotion recognition. For this purpose, a corpus of emotional speech data recorded and classified according to the emotion using the subjective evaluation were used to make statistical feature vectors such as average, standard deviation and maximum value of pitch and energy and to evaluate the performance of the conventional pattern matching algorithms. In the Experiments, pitch and energy parameters were used as a prosodic information and MFCC parameters were used as phonetic information. Experimental results showed that vector quantization based emotion recognizer using MFCC parameters showed better performance than that using the pitch and energy parameters. The vector quantization based emotion recognizer achieved recognition rates of 73.3% for the speaker and context independent classification.

코의 첨부에 발생한 근상피종 : A case report

한원철 圓光大學校 醫科學硏究所 1997 圓光醫科學 Vol.13 No.1-2

Myoepithelioma is a rare tumor of salivary gland, accounting for less than 1 percent of all salivary gland tumors. We experienced a case of myoepithelioma of the nose. Microscopic examination showed characteristic pattern of plasmacytoid type of myoepithelioma.

하수슬러지 중의 중금속

이원준 國立 麗水大學校 環境問題硏究所 2002 環境硏究論文集 Vol.5 No.-

In utilization of sludge as an fertilizer for soil, heavy metal concentration is critical in terms of toxical effects. The heavy metal contents in the sludge can contaminate the soil in long term basis, and the contamination of heavy metal in soil will not be attenuated in nature. This study review the heavy metal standard for sludge used for fertilizer in Korea, Japan, Europe, and America, and the sludge produced in four wastewater plant was investigated. In conclusion the sludge produced from wastewater plant receiving wastewater from industrial complex exceed the Cadmium standard of fertilizer, but other sludge can satisfy the standard. However, the overall heavy metal contents were high enough to consider long term effect.

유전특성에 미치는 선형 저밀도 폴리에틸렌과 에틸렌 비닐 아세테이트 혼합비 영향

김원종,정의남,임윤희,심낙순,박하용 三陟大學校 2002 論文集 Vol.35 No.-

The physical and electrical properties of electrical insulating materials due to the mixing ratio of linear low density polyethylene(LLDPE) and ethylene vinyl acetate(EVA) are studied. Specimens are selected as LLDPE and EVA of mixing ratio of 50 : 50, 60 40, 70 : 30, 80 20 and virgin. To analyze the physical properties of specimens, Fourier transform infra-red spectrum, X-ray diffraction and differential scanning calorimetry are used. To investigate the influence on dielectric characteristic due to Physical structural change of specimens, we investigated the frequency and temperature dependency of tan f for virgin specimen and mixture specimens in the temperature range of 25[t]~120['c], frequency range of 30[Hz] ~l.5×10s[Hz]. For the value of tan 3, dielectric loss, at each temperature of 35, 75, 100, 120[℃], it is confirmed that as applied voltage increase, α and β peak moved to low frequency area at high temperature area, and also moved to high frequency at low temperature areas, the width of peak widen. it is considered to be caused by the molecular vibration, reduction of side-branch density, reduction of crystalline with increasing temperature.

확률적 매칭을 사용한 화자독립 음소 HMM 기반의 음성 인식

김원구 국립7개대학공동논문집간행위원회 2004 공업기술연구 Vol.4 No.-

This paper presents a method that improves the performance of the personal voice dialling system in which speaker independent phoneme HMM's are used. Since the speaker independent phoneme HMM based voice dialing system uses only the phone transcription of the input sentence, the storage space could be reduced greatly.However, the performance of the system is worse than that of the system which uses the speaker dependent models due to the phoneme recognition errors generated when the speaker independent models are used. In order to solve this problem, a new method that jointly estimates transformation vectors for the speaker adaptation and transcriptions from training utterances is presented. The biases and transcriptions are estimated iteratively from the training data of each user with maximum likelihood approach to the stochastic matching using speaker-independent phone models. Experimental result shows that the proposed method is superior to the conventional method which used transcriptions only.

피치 변환 기법을 사용한 실시간 음성 변환 알고니즘

김원구 국립7개대학공동논문집간행위원회 2002 공업기술연구 Vol.2 No.-

In this paper, real-time voice change method using pitch change technique is proposed to change one's voice to the other voice. For this purpose, sampling rate change method using DFT (Discrete Fourier Transform) method and time scale modification method using SOLA(Synchronized OverLap and Add) method is combined to change pitch. In order to evaluate the performance of the proposed method, voice transformation experiments were conducted. Experimental results showed that original speech signal is changed to the other speech signal in which original speaker's identity is difficult to find.