• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10a
• /
• pp.559-562
• /
• 2001

한국어 TTS시스템에서 한 가지 모듈로써의 발음열 생성기는 한국어의 특성상 음운적 조건과 형태론적 조건 등에 의해 다양한 방법과 예외처리를 요구하고 한국어의 음운현상에 대한 체계적인 분석과 처리가 필요하다. 그래서 이 논문은 형태 음운론적 분석을 통한 발음열 자동 생성기법을 소개한다. 이 시스템은 형태소 분석을 선행한 후, 특수문자나 숫자 등을 정규화하고 복합명사 분해 사전을 이용한 복합명사 분해와 추가 조건을 통해 ㄴ-첨가 규칙을 전 처리한다. 그리고 음운 변화 현상을 분석하여 선택적으로 규칙을 적용하여 발음열을 생성한다. 제안된 시스템은 기존의 형태소 분석되지 않은 시스템에 비해 더욱 효과적인 음운, 형태소 변화를 가져옴과 함께, 특히 ㄴ-첨가가 적용되는 텍스트는 7$\sim$8%정도의 나은 발음열을 생성찬 수 있었다. 그 결과, 발음열 생성기는 한국어 TTS 시스템의 한국어 처리라는 고질적인 문제 해결에 좋은 방향과 결과를 기여할 수 있다.

• Education of Primary School Mathematics
• /
• v.13 no.2
• /
• pp.55-66
• /
• 2010

The basic direction of mathematical education for the 21st century is focused on helping student to understand mathematics and developing their problem solving abilities, mathematical disposition and mathematical thinking. Elementary mathematics teachers should help students make sense of mathematics, confident of their ability, and make learning environment comfortable for students to participate in. Through making activities 1 to 100 with digits 1,9,9,6, students improved the interest and preference of students about mathematics. This game is useful to foster students' mathematical thinking(concepts of exponential number expression, roots concept(${\sqrt}$), gauss function([])) and mathematical disposition. If students are helped to be interested in mathematics through mathematical games, they regard mathematics as interesting and challengeable subject to let themselves think many ways.

• Research in Community and Public Health Nursing
• /
• v.7 no.1
• /
• pp.170-177
• /
• 1996

신체의 활력을 측정하는 혈압은 다른 어느 측정치보다 변동이 심하고 정상적으로 변할 수 있는 생리적 현상이어서 정확하게 측정하기란 쉽지 않다. 1mmHg 차이의 혈압일지라도, 특히 큰 집단일 경우, 연구목적에서 그 결과를 크게 왜곡 시킬 수 있다. (Wright & dore, 1970). 혈압 측정에는 직접적 방법과 간접적 방법이 있다. 이 중 간접적 방법은 임상분야와 연구분야에서 실제로 넓게 사용되고 있을 분만 아니라 신체로의 직접적 침투하는 방법도 아니고 가격 또한 비싸지 않다. 그러나, 간접적 측정 방법은 혈압에 영향을 끼치는 많은 요인들인 관찰자 편견, 확장기음의 청취음 선택, 관찰자와 환자 사이의 상호관계 때문에 방법론적으로 문제제기가 되기도 한다. 이런 요인을 최소화하기 위하여 가장 타당도가 높은 측정도구의 선택이 필요하다. 도시지역의(Ease Baltimore) 고혈압을 가진 젊은 흑인 남성(연령별/성별/종속별/거주지역) 중 수축기 혈압이 140mmHg 이고 확장기 혈압이 90mmHg인 사람, 또는 확장기 혈압이 90mmHg인 사람, 또는 혈압하강제를 먹고 있는 사람 200명을 대상으로 훈련받은 community health worker(CHW)에 의해 Random-Zero Sphygmdmanometer(RZS)와 Digit Automatic Device(DAD)로 혈압을 측정하게 하였다. 이 중 36명을 무작위 추출하여 regular Mercury Sphygmdmanometer (RMS)로 혈압을 측정하게 했다. Bland 와 Altman에 의해 제기된 두측정값의 차이에 대한 scattergram(일치도 검증법)과 AAMI(Association for the Advancement of Medical Instrument) 기준, BHS 기준, Central Tendency와 숫자선호도로 검증되었다. AAMI와 BHS 기준에 의하면 확장기 혈압에서 RZS와 DAD의 일치도와 평균을 중심으로 더 넓게 퍼져있어 낮은 일치도를 보였으며 또 DAD와 RMS의 혈압치는 숫자 선호도를 볼 수 있는데 특히 RMS는 더 뚜렷하게 나타났는데 확장기 혈압의 최빈치(30.6)는 100mmHg이었다. 혈압치들이 가장 가까이 "0"점으로 치우쳐져 있었다. RZS의 Central tendency는 DAD와 RMS보다 뚜렷하였다. 결론적으로 RZS는 이 연구목적으로 가장 적절함을 보여준다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.9C
• /
• pp.871-879
• /
• 2009

The performance variations of a spatio temporal smart antenna which is equipped at the basestation of CDMA cellular communication network due to the parametric change of multipath fading environment are studied in this paper. The smart antenna of interest employs space diversity based adaptive array structure in conjunction with rake receiver that has fingers the number of which is the same as that of multipath links. The beamforming is achieved via LMS(Least Mean Square) algorithm in which a reference signal is generated using decision directed formula. It has been shown by computer simulation that the performance of our smart antenna of interest depends significantly upon not only the degree of desired signal's DOA(Direction of Arrival)spread but the number of fingers of the rake receiver. The relative insensitivity of the smart antenna's performance on desired signal's delay spread has also been observed. Computer simulation has shown that the increase of the number of fingers brings in a nonlinear enhancement of the performance of our smart antenna. The renewal of weight vector in the beamforming procedure is taken place at post PN despread stage.

• Journal of the Microelectronics and Packaging Society
• /
• v.27 no.4
• /
• pp.61-65
• /
• 2020

Suppressing lattice thermal conductivity of thermoelectric materials is one of the most popular approach to improve their thermoelectric performance. However, accurate characterization of suppressed lattice thermal conductivity is challenging as it can only be acquired by subtracting other contributions to thermal conductivity from the total thermal conductivity. Here we explain that electronic thermal conductivity (for all materials) and bipolar thermal conductivity (for narrow band gap materials) need to be determined accurately first to characterize the lattice thermal conductivity accurately. Methods to calculate Lorenz number for electronic thermal conductivity (via single parabolic model and using a simple equation) and bipolar thermal conductivity (via two-band model) are introduced. Accurate characterization of the lattice thermal conductivity provides a powerful tool to accurately evaluate effect of different defect engineering strategies.

• Radiation Oncology Journal
• /
• v.10 no.1
• /
• pp.15-20
• /
• 1992

A total of 20 patients (male: female=10:10, adult:children=8:12) with brainstem tumors had been received radiation therapy in the Department of Radiation Oncology, Yonsei Cancer Center, Yonsei University College of Medicine between 1980 and 1990. Thirteen of 20 patients were treated with conventional radiation therapy (before 1989, 180~200 cGy per fraction, 5 days a week, total dose 4680~5400 cGy), and seven patients were treated with hyperfractionated radiation therapy (in 1990, 100 cGy per fraction, twice daily 10 fractions a week, total dose 7200 cGy). Median follow up Periods for conventional radiation therapy group and hyperfractionated radiation therapy group were 36 months and 10 months, respectively. Four of 20 Patients had histopathologic diagnosis prior to treatment; 3 cases were low grade astrocytoma and 1 case was high grade astrocytoma. Overall 2-year actuarial survival rate was 30%. The prognosis of patients with a longer duration of symptom and sign was better (60%, vs 12%), and the adult (52%) was better than children (14%). There was no significant difference between the focal (29%) and diffuse (26%) type. The initial clinical response was better in the hyperfractionated radiation therapy group. Because of the relatively small number of Patients and short follow up Period in hyperfractionated radiation therapy group, there was no comparison between two group.

• Journal of Creative Information Culture
• /
• v.6 no.1
• /
• pp.41-48
• /
• 2020

In order to represent and process spatial data, hierarchical data structures such as a quadtree or a bintree are used. Various approaches for linearly representing the bintree have been proposed. S-Tree has the advantage of compressing the storage space by expressing binary region image data as a linear binary bit stream, but the higher the resolution of the image, the longer the length of the binary bit stream, the longer the storage space and the lower the search performance. In this paper, we construct a hierarchical structure of multiple separated bintrees with a full binary tree structure and express each bintree as two linear binary bit streams to reduce the range required for image search. It improves the overall search performance by performing a simple number conversion instead of searching directly the binary bit string path. Through the performance evaluation by the worst-case space-time complexity analysis, it was analyzed that the proposed method has better search performance and space efficiency than the previous one.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.8
• /
• pp.11-18
• /
• 2009

This paper proposes the way of improving learning speed in Levenberg-Marquardt algorithm using the principal submatrix of Jacobian matrix. The Levenberg-Marquardt learning uses Jacobian matrix for Hessian matrix to get the second derivative of an error function. To make the Jacobian matrix an invertible matrix. the Levenberg-Marquardt learning must increase or decrease ${\mu}$ and recalculate the inverse matrix of the Jacobian matrix due to these changes of ${\mu}$. Therefore, to have the proper ${\mu}$, we create the principal submatrix of Jacobian matrix and set the ${\mu}$ as the eigenvalues sum of the principal submatrix. which can make learning speed improve without calculating an additional inverse matrix. We also showed that our method was able to improve learning speed in both a generalized XOR problem and a handwritten digit recognition problem.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.6
• /
• pp.45-52
• /
• 2023

Recently, as the importance of walking in daily life has been emphasized, projects to guarantee walking rights and create a pedestrian environment are being promoted throughout the region. In previous studies, a pedestrian environment assessment was conducted using Jeonju-si road images, and an image comparison pair data set was constructed. However, data sets expressed in numbers have difficulty in generalizing the judgment criteria of pedestrian environment assessors or visually identifying the pedestrian environment preferred by pedestrians. Therefore, this study proposes a method to interpret the results of the pedestrian environment assessment through data visualization by building a web application. According to the semantic segmentation result of analyzing the walking environment components that affect pedestrian environment assessors, it was confirmed that pedestrians did not prefer environments with a lot of "earth" and "grass," and preferred environments with "signboards" and "sidewalks." The proposed study is expected to identify and analyze the results randomly selected by participants in the future pedestrian environment evaluation, and believed that more improved accuracy can be obtained by pre-processing the data purification process.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.3
• /
• pp.175-182
• /
• 2004

There is a growing interest in speaker verification, which verifies someone by his/her voices. There are many speaker vitrification algorithms such as HMM and DTW. However, it is impossible to apply these algorithms to memory limited applications because of large number of feature vectors to register or verify users. In this paper we introduces a speaker verification system using SVM, which needs a little memory usage and computation time. Also we proposed hardware architecture for SVM. Experiments were conducted with Korean database which consists of four-digit strings. Although the error rate of SVM is slightly higher than that of HMM, SVM required much less computation time and small model size.