• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2008.10a
• /
• pp.324-328
• /
• 2008

금융산업에서 고객의 이탈비율은 기대수익에 영향을 미친다는 점에서 예측이 필요한 부분이며 최근 들어 정확한 예측을 통한 비용관리가 이루어지면서 고객 이탈을 예측하는 것이 중요한 문제로 떠오르고 있다. 그러나 보험 고객 데이터가 대용량이고 불균형한 출력 값을 갖는 특성으로 인해 기존의 방법으로 예측 모델을 만드는 것이 적합하지 않다. 본 연구에서는 대용량 데이터를 처리하는 데 효과적으로 알려져 있는 Trust-region Newton method를 적용한 로지스틱 회귀분석을 통해 이탈고객을 예측하는 것을 주된 연구로 하며, 불균형한 데이터에서의 예측정확도를 높이기 위해 Oversampling, Clustering, Boosting 등을 이용하여 고객 데이터에 적합한 이탈 고객 예측 모형을 제시하고자 한다.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.43 no.6
• /
• pp.906-915
• /
• 1994

Line-Current harmonics resulting from ac to dc power conversion interfere with power system operation and reduce power factor, hence resulting in increasing power source unnecessarily. This study describes a 3 phase phase-controlled converter eliminating low frequency harmonics in the output by inserting notches. Notch points are calculated by Newton-raphson method, The impacts of the choppings on ac and dc sides are analyzed in the steady state. Potential applications of the study are dc motor controls, high power dc power supplies for telecommunications, static var compensators and HVDC.

• Proceedings of the KIEE Conference
• /
• 2000.07a
• /
• pp.80-84
• /
• 2000

Network parameters in power systems are indispensable for all of power system engineering studies, including the power flow calculation and the state estimation. The network parameters required for the studios, in general, are estimated by using several estimation techniques, since it Is very difficult to measure. To improve the estimation accuracy of the network parameters, this paper adopt the synchronized phasor measurements which are acquired from the Phasor Measurement Unit with built-in GPS receiver. In this paper, the parameter estimation problem is formulated with over-determined nonlinear measurement equations and solved with Newton-Raphson method and pseudo-inverse. The effectiveness of the proposed parameter estimation with the synchronized phasor measurements is verified through some case studies with IEEE sample system. The results are very promising.

• Communications for Statistical Applications and Methods
• /
• v.12 no.1
• /
• pp.149-167
• /
• 2005

The minimum disparity estimators introduced by Lindsay and Basu (1994) are studied empirically. An extensive simulation in this paper provides a location estimate of the small sample and supplies empirical evidence of the estimator performance for the univariate contaminated normal model. Empirical results show that the minimum generalized negative exponential disparity estimator (MGNEDE) obtains high efficiency for small sample sizes and dominates the maximum likelihood estimator (MLE) and the minimum blended weight Hellinger distance estimator (MBWHDE) with respect to efficiency at the contaminated model.

• Water for future
• /
• v.29 no.6
• /
• pp.217-224
• /
• 1996

In this paper, the evaluation of diameter, global velocity, global roughness coefficients of the water distribution pipes are examined by using pressure and flowrate measurements in selected points of the network. The selected pipe network parameters are determined through reformulation of the continuity and energy equation. Additional energy equation is considered to analyze the coefficient matrix. The resulting nonlinear equations are solved by using Newton Raphson method. Three computer models with complex pipe system are used to demonstrate these procedures. The computed results of hydraulic parameters show good agreements with KYPIPE2 flow analysis model.

• Tribology and Lubricants
• /
• v.19 no.4
• /
• pp.195-202
• /
• 2003

Many computational researches have been performed about EHL film thickness in the contact between cam and follower in the engine valve train system. However, those computations do not explain the characteristics of dynamic film thickness which means squeeze film effect. Without the consideration of transient term in the Reynold's equation, the predicted film thickness from steady state condition has large difference from the actual film thickness. In this study, we have investigated the kinematic and dynamic simulations of rocker-arm valve train system. From the dynamic simulation, the applied load and the entraining velocity of the lubricant between cam and follower are obtained and with these values the dynamic film thickness is computed by Newton-Raphson method and compared with the steady state film thickness.

• Proceeding of KASS Symposium
• /
• 2008.05a
• /
• pp.113-118
• /
• 2008

This paper presents a flexural-torsional analysis of thin-walled open-section composite beams. A general geometrically nonlinear model for thin-walled composite beams and general laminate stacking sequences is given by using systematic variational formulation based on the classical lamination theory. The nonlinear algebraic equations of present theory are linearized and solved by means of an incremental Newton-Raphson method. Based on the analytical model, a displacement-based one-dimensional finite element model is developed to formulate the problem. Numerical results are obtained for thin-walled composite beams under general loadings, addressing the effects of fiber angle, laminate stacking sequence and loading parameters.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.1089-1090
• /
• 2008

In this paper, it is applied the reconstruction algorithm to detection of breast cancer. To solve the forward and inverse problem, Finite-Difference Time-Domain (FDTD), Tikhonov regularization and Gauss-Newton method are used. And to confirm the accuracy of the algorithm, it is applied to arbitrary model.

• Journal of applied mathematics & informatics
• /
• v.8 no.3
• /
• pp.935-942
• /
• 2001

A finite element approximation of a fourth-order nonlinear boundary value problem is given. In the direct implementation, a nonlinear system will be obtained and also a full size matrix will be introduced when Newton’s method is adopted to solve the system. To avoid this difficulty we introduce an iterative scheme which can be shown to converge the positive solution of the system lying between 0 and $sin{\pi}x$.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2001.04a
• /
• pp.618-621
• /
• 2001

The proposed lever actuator has no friction and mass balance characteristics in motion, which are adapt to high-speed and high-density optical disk system. This paper discussed about the theoretical analysis of the lever structure. The modeling of the lever actuator is found. Using the Newton's method, the motion of equation is deduced through the constraint equations and equilibrium equations in three directions (focusing, tracking and tilting). From the above analysis, we know that the shape of the hinge is the very important parameter on determining the performance of the lever actuator, and the actuator has the 2nd order system characteristics. And the first resonant frequency in transmissibility is dependent to the rigidity of the lever while the first transmissibility resonance of conventional actuators is dependent to the first natural resonance of those actuators. This means that the lever actuator is more stable to the external vibration.