• The Korean Journal of Applied Statistics
• /
• v.29 no.7
• /
• pp.1361-1371
• /
• 2016

Quantile regression provides a variety of useful statistical information to examine how covariates influence the conditional quantile functions of a response variable. However, traditional quantile regression (which assume a linear model) is not appropriate when the relationship between the response and the covariates is a nonlinear. It is also necessary to conduct variable selection for high dimensional data or strongly correlated covariates. In this paper, we propose a penalized quantile regression tree model. The split rule of the proposed method is based on residual analysis, which has a negligible bias to select a split variable and reasonable computational cost. A simulation study and real data analysis are presented to demonstrate the satisfactory performance and usefulness of the proposed method.

• Journal of Ocean Engineering and Technology
• /
• v.23 no.5
• /
• pp.71-78
• /
• 2009

In ship design or offshore structure design, the evaluation of buckling strength (or ultimate strength) is critical to the determination of scantling of stiffened plates. For this reason, it is useful to study the effect of applying different formula or the relationship between stiffened plate with buckling utilization factor (UF). It can facilitate a designer to decide how much the scantling should be reinforced or how much can be reduced for an optimal design. This paper conducts a comparative study for three buckling check methods; DNV-Ship-Rule, DNV-RP-C201, DNV-PULS. The capacity curves and 2D contour plot for utilization factors versus bi-axial in-plane stresses are compared. The contour plots of DNV-Ship-Rule and DNV-PULS show smoothly increasing trends of UF as the applied in-plane stresses increase, however that of DNV-RP-C201 shows rapidly increasing trend as the applied stresses go beyond transverse buckling stress. A sensitivity analysis is performed to investigate the influence level of each parameter of a stiffened plate on UF. Resulting from the analysis, plate thickness is identified to be the most affective parameter to UF regardless of the buckling check methods. Based on the addressed study, optimal designs for bottom plate of 165 K tanker corresponding to three formulas are compared with each other. DNV-PULS yields 1 mm and 2 mm less thickness than DNV-Ship-Rule and DNV-RP-C201, respectively.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2018.05a
• /
• pp.166-168
• /
• 2018

해기사의 행동오류는 해양사고를 야기하는 최종 결과물이다. 해기사의 행동오류는 SRK(Skill, Rule, Knowledge) 행동이론을 통해서 규명할 수 있고, 이러한 행동오류는 해양사고 기록을 통해서 식별할 수 있다. 본 발표의 목적은 과거 해양사고 기록을 토대로 해기사들의 다양한 행동오류와 이들 행동오류로 인하여 발생한 해양사고를 연계하여 해기사들의 행도오류를 평가하기 위한 방법과 현재까지의 결과를 소개하는데 있다. 라스무센이 제한한 행동이론에 해양사고 기록을 적용하여 해양사고 발생 당시의 해기사들의 행동오류를 식별할 수 있었다. 식별한 변수를 이용하여 해기사들의 행동오류를 추정한 결과, 사고원인과 결과 및 행동오류 사이에 강한 상관관계를 나타냈다. 추후, HMM(Hidden Markov Model)을 이용한 예측 기법을 적용하여 행동오류에 따른 해양사고 예방 방안을 제안할 예정이다.

• Journal of Korean Ophthalmic Optics Society
• /
• v.19 no.3
• /
• pp.377-382
• /
• 2014

Purpose: To investigate the effect of anterior, posterior corneal astigmatism and total corneal astigmatism on posterior corneal astigmatism by analyzing correlation. Methods: Participants were 31 patients (31 eyes) without corneal disease at the age range of 22 to 28 who had visited hospital to receive corneal refractive surgery. The total corneal astigmatism and anterior and posterior corneal astigmatism were measured using a rotating scheimpflug camera before surgery. The magnitude of astigmatism was calculated with the difference between the meridian of the steep refractive power and the flat, and With-The-Rule and inverse astigmatism were divided according to the direction of the meridian that was the steepest. Results: The averages of total corneal astigmatism and anterior and posterior astigmatism were found to be $1.13{\pm}0.76D$, $1.51{\pm}0.84D$, and $-0.59{\pm}0.17D$. The magnitude of posterior corneal astigmatism was distributed between -1.0 D and -0.25 D in all the subjects, and when the magnitude of total corneal astigmatism was set as 100, the magnitude of anterior corneal astigmatism was $142.9{\pm}29.9%$. Total corneal astigmatism indicated the highest correlation with the magnitude of anterior astigmatism (y = 0.871x-0.184, $R^2=0.982$) and high negative correlation with posterior astigmatism (y = -2.974x-0.184, $R^2=0.698$). All the subjects' anterior and posterior corneal astigmatism was classified into With-The-Rule. Conclusions: Anterior and posterior corneal astigmatism in the subjects in their 20's showed the magnitude of -3.8 D in 0.2 D and -0.25 D in -1.0 D separately, and both anterior and posterior astigmatism indicated a high percentage of With-The-Rule.

• Korean Journal of Food Science and Technology
• /
• v.33 no.6
• /
• pp.700-705
• /
• 2001

Sucrose첨가가 옥수수 전분호액의 레올로지 특성에 미치는 영향에 대하여 평가하였다. 전분호액은 sucrose 첨가에 관계없이 전단담화 비뉴턴 거동(n=0.37-0.58)을 나타내었으며, 또한 sucrose의 농도가 증가함에 따라 점조도지수$(K,\;K_h)$, 겉보기 점도$({\eta}_{a,100})$값은 크게 감소하였고 항복응력$({\sigma}_{oc},\;{\sigma}_h)$도 감소하는 경향을 나타내었다. 전분호액의 겉보기점도에 대한 온도의존성은 Arrhenius식에 의하여 높은 상관관계를 나타내었으며, 활성화에너지는 각각 10.80-15.59 kJ/mole을 나타내었고 농도가 증가함에 따라 감소하는 경향을 나타내었다. 적용된 주파수$({\omega})$ 범위에서 저장탄성률(G#)의 수치가 손실탄성률(G@)보다 높게 나타났으며 sucrose의 농도가 증가함에 따라 감소하여 탄성 및 점성 성질을 감소시켜 주었다. 이들 동적데이터로부터 전분호액은 sucrose 첨가에 의해 더욱 약한 겔과 같은 구조적 성질을 나타냄을 알 수 있었다. 그러므로 sucrose 첨가는 전분호액 내의 입자들의 재회합을 방해하여 전분호액의 레올로지 특성을 변화시킨다. Sucrose 30%를 첨가한 옥수수 전분호액을 제외한 모든 시료들은 이동인자$(shift\;factor,\;\alpha)$를 사용함으로써 Cox-Merz 중첩 원리에 잘 적용되었다.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.14 no.2
• /
• pp.97-112
• /
• 1978

The rule 15, 16 and 17 of International Regulations for Preventing collisions at Sea direct actions to avoid collision when two power-driven vessels are crossing. But these rules do not present the safety minimum approaching distances outside which a give- way vessel deeps out of the way of a stand-on vessel. In this paper, the author analyzed the ship's collision avoiding actions from a viewpoint of ship kinematics as the method to calculate this distance. The author worked out mathematic formulas for calculating the safety minimum approaching distances outside which the give-way vessel takes the actions to avoid collisions in accordance with the cross angles of the crossing vessels' courses. Figuring out actually the values of maneuvering indices of the M. S. Koan Ack San (GT: 224tons), the training ship of the National Fisheries University of Busan and the M. S. Golden Clover (GT: 101, 235tons) of the Eastern Shipping Co., Ltd. through their Z test, the author applied these values to the calculating formulas and calculated the safety minimum approaching distances. The results of calculations are as follows; 1. The greatest distance is to be kept by the give-way vessel to avoid collision when the cross angle of courses is 90$^{\circ}$ or near it. In such case the safety minimum approaching distance of a small vessel must be more than 5 times of her own length and that of a large vessel more than 11 times of her own length. 2. Collision danger is greater when crossing angle is obtuse than in an acute angle, therefore greater distance is to be kept by the give-way vessel to avoid collision in the case of the obtuse angle. 3. The actions to be taken to avoid collisions by the give-way vessel in Rule 16 and by the stand-on vessel in Rule 17(a)(ii) of International Regulations for Preventing Collisions at Sea, must be done outside the above safety minimum approaching distance. When inevitably such actions are to be taken within the safety minimum approaching distance, they should be accompanied with engine motions.

• Journal of KIISE:Software and Applications
• /
• v.30 no.5_6
• /
• pp.540-549
• /
• 2003

In collaborative recommendation two models are generally used: the user model and the article model. A user model learns correlation between users preferences and recommends an article based on other users preferences for the article. Similarly, an article model learns correlation between preferences for articles and recommends an article based on the target user's preference for other articles. In this paper, we investigates various combination methods of the user model and the article model for better recommendation performance. They include simple sequential and parallel methods, perceptron, multi-layer perceptron, fuzzy rules, and BKS. We adopt the multi-layer perceptron for training each of the user and article models. The multi-layer perceptron has several advantages over other methods such as the nearest neighbor method and the association rule method. It can learn weights between correlated items and it can handle easily both of symbolic and numeric data. The combined models outperform any of the basic models and our experiments show that the multi-layer perceptron is the most efficient combination method among them.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.6
• /
• pp.108-114
• /
• 1989

Introducing and optimizing bit-significance to the Hopfield model, ten highly correlated binary images, i.e., numbers "0" to "9", are successfully stored and retrieved in a $6{}8$ node system. Unlike many other neural network models, this model has stronger error correction capability for correlated images such as "6","8","3", and "9". The bit significance optimization is regarded as an adaptive learning process based on least-mean-square error algorithm, and may be implemented with Widrow-Hoff neural nets optimizer. A design for electro-optic implementation including the adaptive optimization networks is also introduced.

• Journal of Korea Water Resources Association
• /
• v.33 no.5
• /
• pp.593-601
• /
• 2000

Analyzing monthly inflows of the Chung-Ju Dam associated with EI Nino Southern Oscillation (ENSO), Kim and Lee(2000) reported that the fall and winter inflows in EI Nino years tended to be low while those in La Nina years tended to be high. This study proposes a methodology of employing such a teleconnection between ENSO and inflow in reservoir operations. The ENSO information is used as a hydrologic state variable in stochastic dynamic programming (SDP) to derive a monthly optimal rule for operating the Chung- Ju Dam. An alternative operating rule is also derived with the SDP with no hydrologic state variable. Both of the SDP operating rules are simulated and compared to examine the value of using the ENSO information in operations of the Chung-Ju Dam. The simulation results show that the operating rule using the ENSO information increases energy generation and reliability of water supply as well as reduces spill. spill.

• Proceedings of the Korea Water Resources Association Conference
• /
• 1994.02a
• /
• pp.361-366
• /
• 1994

저수지 운영방안 정책결정에 있어 보다 효과적으로 적용할 수 있는 방법론의 개발이 여러측면에서 이루어져 왔다. 이중 동적계획기법의 Explicit 및 Implicit 해에 의한 최적운영방안의 검토가 한강수계의 충주댐을 대상으로 이루어졌다. 이들 방법은 한정된 과거 기록치로부터 합성유량을 발생하여 동적계획기법에 의한 충주댐 최적 운영모형에 적용하여 얻어진 상태변수 및 결정변수의 상관관계를 기준으로 도출한 운영율에 기초하여 모의운영모형을 개발할 수 있다. 개발된 모형중 Explicit 기법은 조건확율에 따른 전단계의 이산화된 유입량 조건별 운영단계의 월초저류량을 기준으로 월말 저류량은 결정하는 방법이며, Implicit 기법은 전단계 저류량 및 유입량, 운영단계 저류량 및 유입량을 대상으로 조합에 의한 회귀분석후 상관성이 우수한 운영율 방정식을 개발하게 된다. 본 연구에서는 이렇게 개발된 두가지 운영율을 기준으로 다목적 운영정책 결정을 위한 저수지 모의운영 모형을 개발하여 모형의 이행도를 평가하였다. Explicit 및 Implicit 기법에 기초한 모의모형의 평가방법은 모의치와 과거 운영실적을 비교하는 것으로 하고 Explicit 기법의 적용에서 홍수기 수문사상의 불확실성에 따른 저수지 운영 효율개선을 위하여 수정 방류량 결정방법을 도입하여 가장 적절한 저수지 운영모의모형 개발방법을 제시하고 있다.