http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
터치 기반의 모바일 폰을 위한 쇼핑몰 사이트의 UX 디자인 분석
최동우(Choi Dong-Woo),이상선(Yi Sangsun) 한국디자인학회 2010 한국디자인학회 학술발표대회 논문집 Vol.2010 No.10
In recent years, the data tfic of mobile web has been increasing dramatically due to growth in the smartphone market. Smartphone users are facing troubles as peculiar environment of mobile and constraints of the devices. Especially, internet shopping sites are over to mobile environment from personal computing environment are undergoing many problems because of complex structure of shopping sites and small screen. I analyzed the mobile shopping sites' UX design to solve these problems. Analysis was based on the W3C's Mobile Web guidelines and iOS ill guidelines. There were various problems which have not been optimized for mobile environment. According to the result, mobile webs having comparatively limited amount of information that is able to be displayed was the major problem.
필터 분해 기법을 이용한 에너지 효율적 재구성형 CNN 가속기 구조
최동우(Dong Woo Choi),이한호(Hanho Lee) 대한전자공학회 2020 전자공학회논문지 Vol.57 No.7
본 논문은 에지 디바이스에서의 CNN 알고리즘 처리를 위한 Row Stationary (RS) 데이터 흐름과 Spatial Reduction (SR) 구조기반 재구성형 Convolutional Neural Network (CNN) 가속기 구조 설계 방법을 제안한다. 제안된 필터 분해 기법을 이용한 에너지 효율적 재구성형 CNN 가속기 구조는 높은 에너지 효율뿐만 아니라 기존 RS 데이터 흐름 기반 CNN 가속기 보다 향상된 데이터 처리율을 보여준다. 제안된 에너지 효율적 재구성형 CNN 가속기는 필터 분해기법을 적용하여 적은 면적으로 다양한 크기의 필터를 대상으로 한 컨볼루션 연산이 가능하며, 138.7 Giga Operation per Second (GOPS)의 데이터 처리율과 30.02 GOP/J의 에너지 효율을 갖는다. 제안된 필터 분해 기법을 이용한 에너지 효율적 재구성형 CNN 가속기는 FPGA Virtex-7을 사용하여 검증하였으며, 다른 FPGA 기반 CNN 가속기 대비 187%∼247% 향상된 에너지 효율과 기존 RS 데이터 흐름 기반 CNN 가속기 대비 390% 향상된 데이터 처리율을 갖는다. 따라서 제안된 구조는 저전력 구동을 요구하는 에지 디바이스에서 CNN 연산을 위해 사용할 수 있다. This paper shows a Row Stationary (RS) dataflow and Spatial Reduction (SR) structure based reconfigurable Convolutional Neural Network (CNN) accelerator architecture for edge device. The proposed energy-efficient reconfigurable CNN accelerator using filter decomposition technique shows improved data throughput over existing RS dataflow-based CNN accelerator as well as high energy efficiency. The proposed energy-efficient reconfigurable CNN accelerator can compute convolution operation with different sizes of filter in low area. The data throughput of proposed energy-efficient reconfigurable CNN accelerator is 168.7 Giga Operation per Second (GOPS) and the energy-efficiency is 35.14 GOP/J. The proposed energy-efficient reconfigurable CNN accelerator using filter decomposition techniques was implemented on FPGA Virtex-7. It has a 187 to 247% improvement in energy efficiency compared to other FPGA-based CNN accelerators and a 251% improvement in data throughput compared to existing RS data flow-based CNN accelerators. Thus, the proposed structure can be used for CNN computation in edge device that require low-power driving.
김윤경 ( Yunkyung Kim ),최동우 ( Dong-woo Choi ),박은철 ( Eun-cheol Park ) 한국보건행정학회 2019 보건행정학회지 Vol.29 No.1
Catastrophic health expenditure refers to spending more than a certain level of household’s income on healthcare expenditure. The aim of this study was to investigate the proportion of households that experienced catastrophic health expenditure between 2006 and 2017 with the National Survey of Tax and Benefit (NaSTaB) and between 2011 to 2016 using Households Income and Expendi-ture Survey (HIES) data. The results of the NaSTaB showed 2.16% of households experienced the catastrophic health expenditure in 2017. In trend analysis, the NaSTaB revealed a statistically significant decreasing trend (annual percentage change [APC]=-2.01, p<0.001) in the proportion of households with the catastrophic health expenditure. On the other hand, the results of the HIES showed 2.92% of households experienced the catastrophic health expendi-ture in 2016. Also, there was a slightly increasing trend (APC=1.43, p<0.001). In subgroup analysis, groups with lower income levels were likely to experience catastrophic health expenditure. In conclusion, further public support system is needed to lower experience these healthcare expenditures and monitor the low income group.