• 한국응용곤충학회지
• /
• 제32권1호
• /
• pp.42-50
• /
• 1993

우리나라에 분포하고 있는 재래꿀벌(Apis cerana)의 일벌을 대상으로 춘계 11지역 및 하계 12지역 개체군을 선발하였으며, 총 47개 정량형질에 대해 지역별로 판별분석을 실시 하였고, 또한 양 계절간 변이도 분석하였다. 그 결과, 지역별 비교에서는 춘계 91.67%, 하계 88.44%의 평균분리율을 보였으며, 양계절간 비교에서는 97.58%로 나타났다. 반면에 양계절을 통합한 비교에서는 70.16%로 낮았다.

• 아동학회지
• /
• 제23권5호
• /
• pp.89-104
• /
• 2002

This investigation compared the eyewitness accounts of 5-year-old children by verbal-only inter views and interviews using anatomically neutral dolls. The verbal interviews consisted of free recall, specific question, and leading questions. While the overall accuracy of the interviews increased with the introduction of dolls as memory aids, the efficacy of the dolls was not uniform across types of interviews. In free recall, the dolls were not effective in eliciting accurate accounts. The use of dolls also did not compromise the memory of children in free recall which recalled greater number of correct details with the memory aids than without, but it should be careful in free recall could be reported exactly correct in spite of little bit amounts. Responses to leading questions showed that the children were affected by suggestive misleading question and were susceptible to incorporation of that information into their memory.

• 음성과학
• /
• 제14권1호
• /
• pp.151-162
• /
• 2007

This study focuses on the classification of pathological voice using GMM (Gaussian Mixture Model) and compares the results to the previous work which was done by ANN (Artificial Neural Network). Speech data from normal people and patients were collected, then diagnosed and classified into two different categories. Six characteristic parameters (Jitter, Shimmer, NHR, SPI, APQ and RAP) were chosen. Then the classification method based on the artificial neural network and Gaussian mixture method was employed to discriminate the data into normal and pathological speech. The GMM method attained 98.4% average correct classification rate with training data and 95.2% average correct classification rate with test data. The different mixture number (3 to 15) of GMM was used in order to obtain an optimal condition for classification. We also compared the average classification rate based on GMM, ANN and HMM. The proper number of mixtures on Gaussian model needs to be investigated in our future work.

• 한국언어정보학회지:언어와정보
• /
• 제4권2호
• /
• pp.69-92
• /
• 2000

It is well-known that default rules require a nonmonotonic logic. Veltman proposed one dynamic theory which interprets default rules in such a way that correct inferences can be made at each information state. But his theory has some problems. First, this theory excludes the possibility that a default rule can be true of false. Second, his representation of an information state makes it difficult to interpret a default rule embedded in another sentence. Third, the notion of a frame which is introduced in the interpretation of a default rule and the adjustment of inferential expectation has a more complex structure than is necessary, In this paper, I propose a truth-conditional theory of default rules in which the meaning of a default rule is defined as a truth-condition in a possible world and which assumes a simpler structure of a frame. This makes it possible to interpret a default rule embedded in a sentence. A dynamic theory for default rules is also proposed for correct inferences based on default rules.

• 대한방사선기술학회지:방사선기술과학
• /
• 제22권1호
• /
• pp.35-42
• /
• 1999

This study was undertaken to take the correct vocal technique(especially about the resonance of oral cavity). The resonance of oral and pharyngeal cavity is the principle which can vocalize well without any abnormal signs in the throat. Therefore it is important for us to understand how to use the correct resonance of oral and pharyngeal cavity. Shimadzu X-ray remote control TV system and Shimadzu magnet $nex-{\alpha}$ (SMT-50CX/H) were used for checking the movements of T-M joint and diaphragmatic respiration. The results obtained were summerized as follows: 1. While opening T-M joint space like the vowel "A" [a], We should vocalize five fundamental vowel [a,e,i,o,u] with diaphragmatic respiration holded. 2. Diminuendo must be expressed by increasing a breath volume while descending a mandible gradually because we can not ascend maxilla. So we can make a delicate expression. 3. The resonance of oral cavity must be scattered by elevating the soft palatine lightly with relax of throat.

• 응용통계연구
• /
• 제21권1호
• /
• pp.169-176
• /
• 2008

일반화선형모형 에서 추가되는 설명 변수의 비선형성의 존재와 형태를 파악하는데 사용되는 3차원 CBRBS그림의 구조와 유용성에 대해 알아본다. CBRBS그림은 설명변수들 사이에 비선형의 관계가 존재하는 경우 편잔차그림으로는 알아낼 수 없는 비선형성에 대한 탐지가 가능하다. 이를 생성된 자료를 이용하여 확인해 본다.

• 기술경영경제학회:학술대회논문집
• /
• 기술경영경제학회 1997년도 제12회 동계학술발표회 논문집
• /
• pp.197-228
• /
• 1997

This study aims to evaluate and identify the Patterns of the technological competitiveness of the Korean industry. Such statistics as R&D expenditure and R&D manpower as input indexes, US patent registrations and export sales as output indexes were used. it was turned out that such industrial types as specialized-suppliers industries, scale-intensive industries and science-intensive industries showed relatively strong technical competitiveness. However, resource-intensive industries and labor-intensive industries which had maintained a competitive advantage in the 1970s and the 1980s appeared to be gradually losing their technological competitiveness. These results are by and large in accordance with the trends of export performance. This study conducted the canonical discriminant analysis in order to test the correctness of the patterns displayed in the technological competitiveness of the Korean industry. The result of the analysis showed that the five patterns of technical strength of the Korean industries are significantly independent each other for four respective variables which are used to distinguish industries. This implies that the ex ante industrial classification into five typers was correct in terms of the ex post statistics, and that the patterns of technological competitiveness discovered in this study are also statistically correct.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2004년도 추계학술대회 논문집
• /
• pp.696-699
• /
• 2004

A side scan sonar system uses the towfish installed sonars, It is an equipment that search images of the bottom surface of the sea in real time. It is a typical equipment that is related to a sea investigation such as a geological survey, seabed communication cable and power line cable placing repair investigation, fish breeding ground investigation, sea purification, relic and mineral investigation, and mine and submarine search. It used to fined objects and investigate on the seabed surface. But, recently, it is used to sea purification and geological survey that require information of the correct surface of the seabed. So, it needs various filtering technique and image processing techniques development to acquire high resolution image. therefore, this research develops a side scan sonar using multi-beam sensors that supply various information with the fast scan speed and correct high resolution that is not a simple underwater investigation equipment.

• 대한음성학회지:말소리
• /
• 제41호
• /
• pp.1-18
• /
• 2001

The purpose of this paper is to slove problems with English pronunciation education and to correct the misbelief that misbelief that English pronunciation can be thught effectiverly by native speakers or good speakers. Best sellers have many mistakes even in the description of sounds and most English education departments in national universities teach English phonetics theoretically only for 1 semester, More than 10 years' investigation into native speakers' and good speakers' teaching methods in various ways confirmed that they couldn't know English phonetics teachers should have a theoretical and practical knowledge of phonetics of English and students' dialects and the abilities to find the same or similar English sounds their students can make in normal or abnormal situations, to make both correct sounds and incorrect ones their students can make and to make them perceive the differences by using various ways.

• 한국기계가공학회지
• /
• 제9권2호
• /
• pp.20-25
• /
• 2010

A camera module is supported by a gimbal structure in collection equipment of image information. During flight, the gimbal system undergoes serious accelerations with wide frequencies. To get the correct images, the camera module must be stably vibrated under these conditions. If natural frequency unfortunately exists in a exciting frequency range, resonance occurs there. Hence, harmonic responses analysis is needed to know correct vibration characteristic of the gimbal system. Finite element analysis was performed to get an acceleration of the gimbal system by mode superposition after extracting mode shapes and natural frequencies. Considering damping ratio of 2%, the reponses of gimbal structure were calculated from excitations with a design frequency band. As results, a maximum acceleration transmissibility, which is the ratio of response to excitation, was obtained and it can be used to design the gimbal structure effectively.