• Communications for Statistical Applications and Methods
• /
• v.16 no.1
• /
• pp.185-193
• /
• 2009

Machine learning methods such as support vector machines and random forests yield nonparametric prediction functions of the form y = $f(x_1,{\ldots},x_p)$. As a sequel to the previous article (Huh and Lee, 2008) for visualizing nonparametric functions, I propose more sensible graphs for visualizing y = $f(x_1,{\ldots},x_p)$ herein which has two clear advantages over the previous simple graphs. New graphs will show a small number of prototype curves of $f(x_1,{\ldots},x_{j-1},x_j,x_{j+1}{\ldots},x_p)$, revealing statistically plausible portion over the interval of $x_j$ which changes with ($x_1,{\ldots},x_{j-1},x_{j+1},{\ldots},x_p$). To complement the visual display, matching importance measures for each of p predictor variables are produced. The proposed graphs and importance measures are validated in simulated settings and demonstrated for an environmental study.

• Korean Journal of Food Science and Technology
• /
• v.21 no.5
• /
• pp.613-618
• /
• 1989

Tirteen rice varieties of Japonica-Indica briefing type (J-I breeding type) and ten of Japonica type produced in Korea in 1985 were studied on cooking quality of milled rice and texture of cooked rice. The results are as follows: in the Looking quality, water uptake ratio of cooked rice and total solid in residual liquid showed significant $(p<2.172^*\;and\;2.419^*,\;respectively)$ between J-I breeding type and Japonica type The water uptake ratio of cooked rice was higher J-I breeding type than Japonica type but the total solid and intensity of iodine blue value in residual liquid were lower J-I breeding type than Japonica type. Minimum cooking time of milled rice showed significant $(p<2.375^*)$ between J-I breeding type and Japonica type for texture of cooked rice. The cooking time of J-I breeding type shorter than Japonica type. Hardness and cohesiveness in texture of cooked rice appeared to be significant cant $(p<4.370^{***}\;and\;2.371^*,\;respectively)$ between J-I breeding type and Japonica type, and then hardness was higher Japonica type than J-I breeding type, but cohesiveness was lower Japonica type than J-I breeding type. The correlation coefficients between amylose content and adhesiveness of varieties in the same type was $-0.922^{**}$ in J-I breeding type and $-0.915^**}$ in Japonica type. The correlation coefficients between protein content and hardness of the varieties in the same type was $0.954^{**}$ in Japonica type and so denoted as positive correlation. And the scatter diagram between protein and hardness could be divided into J-I breeding type and Japonica type.

• 한국전산유체공학회:학술대회논문집
• /
• 2008.08a
• /
• pp.76.6-77
• /
• 2008

• Transactions of Materials Processing
• /
• v.20 no.1
• /
• pp.23-28
• /
• 2011

For the fabrication of nano patterned products manufacturing a nano patterned mold is needed in advance. The nano patterned stamper was fabricated by electroforming the AAO master with nickel. The surface of nickel-plated stamper had nano-patterned holes with the diameter of 73 nm and the depth of 83 nm. Hot embossing was used for forming P3HT sheet and the process factors of hot embossing were closer as pressure, temperature and time. In the present paper hot embossing experiments were performed to find the main process conditions to affect the replication ratio of nano patterns on surface of P3HT sheet. As a result, main contributing factors for the replication ratio of hot embossed pattern could be sequentially enumerated as pressure, temperature and time.

• Journal of the Korean institute of surface engineering
• /
• v.34 no.5
• /
• pp.414-420
• /
• 2001

A 30-kV plasma immersion ion implantation setup (P $I^3$) has been equipped with a self-developed 6'-magnetron to perform hard coatings with enhanced adhesion by P $I^3$D(P $I^3$ assisted deposition) process. Using ICP source with immersed Ti antenna and reactive magnetron sputtering of Ti target in $N_2$/Ar ambient gas mixture, the TiN films were prepared on Si substrates at different pulse bias and ion-to-atom arrival ratio ( $J_{i}$ $J_{Me}$ ). Prior to TiN film formation the nitrogen implantation was performed followed by deposition of Ti buffer layer under A $r^{+}$ irradiation. Films grown at $J_{i}$ $J_{Me}$ =0.003 and $V_{pulse}$=-20kV showed columnar grain morphology and (200) preferred orientation while those prepared at $J_{i}$ $J_{Me}$ =0.08 and $V_{pulse}$=-5 kV had dense and eqiaxed structure with (111) and (220) main peaks. X-ray diffraction patterns revealed some amount of $Ti_{x}$ $N_{y}$ in the films. The maximum microhardness of $H_{v}$ =35 GN/ $M^2$ was at the pulse bias of -5 kV. The P $I^3$D technique was applied to enhance wear properties of commercial tools of HSS (SKH51) and WC-Co alloy (P30). The specimens were 25-kV PII nitrogen implanted to the dose 4.10$^{17}$ c $m^{-2}$ and then coated with 4-$\mu\textrm{m}$ TiN film on $Ti_{x}$ $N_{y}$ buffer layer. Wear resistance was compared by measuring weight loss under sliding test (6-mm $Al_2$ $O_3$ counter ball, 500-gf applied load). After 30000 cycles at 500 rpm the untreated P30 specimen lost 3.10$^{-4}$ g, and HSS specimens lost 9.10$^{-4}$ g after 40000 cycles while quite zero losses were demonstrated by TiN coated specimens.s.

• Proceedings of the Korean Magnestics Society Conference
• /
• 2009.05a
• /
• pp.186-188
• /
• 2009

• 대한생식의학회:학술대회논문집
• /
• 2001.06a
• /
• pp.45.2-46
• /
• 2001

• 대한생식의학회:학술대회논문집
• /
• 2001.06a
• /
• pp.80.1-80.1
• /
• 2001

• Proceedings of the KIPE Conference
• /
• 2012.11a
• /
• pp.61-62
• /
• 2012

본 논문에서는 10kW급 리튬계열 에너지 저장장치 (ESS;Energy storage system)의 전기화학적 온도특성을 분석하였다. 70개 $LiFePO_4$ 단위 셀의 직렬결합(70S1P)으로 이루어진 ESS의 온도특성 분석을 위해 여름(7월)과 겨울(1월)일때의 ESS Tray/Rack/Cell 전압(표준편차) 및 온도데이터(최대값/최소값)의 통계처리를 실시하였다. 실험결과를 통해, 주변 외기온도에 영향을 최소화한 안정적인 ESS의 최적구축을 위한 가이드 라인을 제시할 수 있다.

• Electronics and Telecommunications Trends
• /
• v.8 no.3
• /
• pp.163-183
• /
• 1993

NTT는 1980년에 DDX-P 패킷통신서비스를 개시하여 현재 DDX-TP, DDX-TP 및 INS-P의 3종류의 서비스를 제공하고 있다. 본 고는 NTT의 패킷통신망 구조, 기능 및 싱호통신 등 NTT 패킷통신서비스 전반적인 면과 이를 이용한 사례를 분석한 결과이다.