• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.6
• /
• pp.545-551
• /
• 2017

It is very important to effectively obtain the information related to the human body shape for user-centered design. The human body shape is a huge combination of various irregular curves and is typically obtained by a 3-D Scanner. 3-D scanners show high reliability; however, they are expnsive equipment with limited mobility. 3-D models of irregular-shaped curves were created by a photogrammetric approach and the errors between the original curve and the models were evaluated. 3-D models were created based on 160, 80, 40, 20, 10, and 5 marking points evenly located on the original curve. In the case of convex curve, low levels of residuals were observed in the models from 160, 80, 40, and 20 marking points (0.13% max). In the combination of convex and concave curves, relatively low levels of residuals were observed in the models from 160, 80, and 40 marking points (0.29%). It is possible to conclude that marking points should be placed at every 5% of overall length of a convex curve and at every 2.5% of overall length of a curve with convex and concave curve in order to maintain low levels of errors. A photogrammetric approach can be used as an alternative for the 3-D scanners with advantages of low cost and mobility.

• Journal of the Korean Society for Precision Engineering
• /
• v.24 no.12
• /
• pp.20-28
• /
• 2007

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.28 no.6
• /
• pp.707-714
• /
• 2015

In this paper, an experimental program has been conducted to investigate transfer length in high strength concrete members pretensioned through a seven-wire strand with FBG sensors. To measure transfer length, five members were fabricated, which had a length of 3 m and a cross-section of $150{\times}150mm$. It was measured that the concrete compressive strength was 58MPa at pretensioning. Test results indicated that more precise and reliable measurement on the transfer length was attained with FBG sensors than conventional gauges attached on concrete surface. Through comparing the measured transfer length and predictions, applicability of several transfer length models in literature was investigated. This paper can be useful for relevant research field such as investigation on the bond mechanism of a seven-wire strand in concrete members.

• Journal of the Korea Concrete Institute
• /
• v.20 no.6
• /
• pp.785-796
• /
• 2008

Longitudinal elongation develops in reinforced concrete beams that exhibit flexural yielding during cyclic loading. The longitudinal elongation can decrease the shear strength and deformation capacity of the beams. In the present study, nonlinear truss model analysis was performed to study the elongation mechanism of reinforced concrete beams. The results showed that residual tensile plastic strain of the longitudinal reinforcing bar in the plastic hinge is the primary factor causing the member elongation, and that the shear-force transfer mechanism of diagonal concrete struts has a substantial effect on the magnitude of the elongation. Based on the analysis results, a simplified method for evaluating member elongation was developed. The proposed method was applied to test specimens with various design parameters and loading conditions.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.2
• /
• pp.45-52
• /
• 2020

Since the opening of the national petition site, it has attracted much attention. In this paper, we perform topic analysis of the national petition site and propose a prediction model for answerable petitions based on deep learning. First, 1,500 petitions are collected, topics are extracted based on the petitions' contents. Main subjects are defined using K-means clustering algorithm, and detailed subjects are defined using topic modeling of petitions belonging to the main subjects. Also, long short-term memory (LSTM) is used for prediction of answerable petitions. Not only title and contents but also categories, length of text, and ratio of part of speech such as noun, adjective, adverb, verb are also used for the proposed model. Our experimental results show that the type 2 model using other features such as ratio of part of speech, length of text, and categories outperforms the type 1 model without other features.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.2
• /
• pp.303-307
• /
• 2006

In this paper, we have designed the p-type metal-oxide semiconductor field-effect transistor(pMOSFET) for SiGe devices with gate lengths of $0.9{\mu}m$ and $0.1{\mu}m$using the TCAD simulators. The electrical characteristics of devices have been investigated over the temperatures of 300 and 77K. We have used the two carrier transfer models(hydrodynamic model and drift-diffusion model). We how that the drain current is higher in the hydrodynamic model than the drift-diffusion model. When the gate length is $0.9{\mu}m$, the threshold voltage shows -0.97V and -1.15V for 300K and 77K, respectively. The threshold voltage is, however, nearly same at $0.1{\mu}m$ for 300K and 77K.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.22 no.2
• /
• pp.115-124
• /
• 2018

In this study, a cold flow test was carried out on a high-speed vehicle facility with a high-altitude environment simulator. Variable test was carried out according to the blockage ratio, angle, and length of the test model. It is confirmed that the blockage rate can be operated in the range of 40%, and that the model should be selected at an angle of 45 degrees or less. The variables of length are less dominant compared to the variables of blockage rate and angle. Through this, a database is obtained according to the parameters of the conical model of the high-speed vehicle test facility.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.04a
• /
• pp.615-618
• /
• 2011

본 논문은 IPTV, VoD와 같은 대용량의 비디오 스트리밍 트래픽을 처리함에 있어 큐 길이 변화를 최소화할 수 있는 SCQ(Smoothly Changing Queue) 스케줄링 방안을 제안한다. SCQ는 벌크하게 유입되는 스트리밍 트래픽에 대해 유입되는 패킷 양 및 속도를 고려하여 서비스율을 제어함으로써 큐 길이 변화율을 최소화할 수 있다. 벌크 특성을 갖는 스트리밍 서비스 전달에 있어 종단간 낮은 지연변이를 유지할 수 있으며 제안 방안을 이용할 경우 보다 안정적으로 서비스를 제공할 수 있다. 제안 방안의 유효성을 확인하기 위해 우리는 큐잉 모델을 이용하였으며, 그 결과 기존 평균 큐길이 기반 스케줄링 방안보다 낮은 큐길이 변화율을 보임을 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.16-18
• /
• 2023

본 연구는 궤적 데이터(trajectory data)를 대상으로 증강 그래프 기반의 그래프 뉴럴 네트워크를 활용하여 다음에 방문한 장소를 추천하는 모델을 제안한다. 제안 모델은 전체 궤적 데이터를 그래프로 표현하여 추출한 글로벌 궤적 플로우의 특성을 다음 방문할 POI 추천에 활용한다. 이때, POI 추천시 자주 발생하는 두 가지 문제를 추가로 해결함으로써 POI 추천의 정확도를 높이는 것을 목표로 한다. 첫 번째 문제는 추천 대상 궤적 데이터의 길이가 짧은 경우에 성능 저하가 발생한다는 것이다. 두 번째 문제는 콜드-스타트 문제이다. 기존 POI 추천 모델은 매우 적은 방문 기록만 가지는 사용자 또는 POI에 대해서는 매우 낮은 예측 성능을 보인다. 본 연구에서는 궤적 그래프에서 일부 엣지를 삭제하여 생성한 증강 그래프 기반의 궤적 플로우 특징 기반 모델을 제안함으로써 짧은 길이의 궤적 데이터 및 콜드-스타트 사용자/POI에 대한 추천 성능을 높인다.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.495-500
• /
• 2020

뉴스 기사는 반드시 객관적이고 넓은 시각으로 정보를 전달하지 않는다. 따라서 뉴스 기사를 기존의 추천 시스템과 같이 개인의 관심사나 사적 정보를 바탕으로 선별적으로 추천하는 것은 바람직하지 않다. 본 논문에서는 최대한 객관적으로 다양한 시각에서 비슷한 사건과 인물에 대해서 판단할 수 있도록 유사도 기반의 기사 추천 모델을 제시한다. 길이가 긴 문서 사이의 유사도를 측정하기 위해 GPT2 [1]언어 모델을 활용했다. 이 과정에서 단방향 디코더 모델인 GPT2 [1]의 단점을 추가 학습으로 개선했으며, 저장 공간의 효율과 핵심 문단 추출을 위해 BM25 [2]함수를 사용했다. 그리고 준 지도 학습 [3]을 통해 유사도 레이블링이 되어있지 않은 최신 뉴스 기사에 대해서도 자가 학습을 진행했으며, 이와 함께 길이가 긴 문단에 대해서도 효과적으로 학습할 수 있도록 문장 길이를 기준으로 3개의 단계로 나누어진 커리큘럼 학습 [4]방식을 적용했다.