• Journal of Intelligence and Information Systems
• /
• v.17 no.3
• /
• pp.99-114
• /
• 2011

Collaborative Filtering (CF) suffers from two major problems:sparsity and cold-start recommendation. This paper focuses on the cold-start problem for new customers with no purchase records and the sparsity problem for the customers with very few purchase records. For the purpose, we propose a method for the new customer recommendation by using a combined measure based on three well-used centrality measures to identify the customers who are most likely to become neighbors of the new customer. To alleviate the sparsity problem, we also propose a hybrid approach that applies our method to customers with very few purchase records and CF to the other customers with sufficient purchases. To evaluate the effectiveness of our method, we have conducted several experiments using a data set from a department store in Korea. The experiment results show that the combination of two measures makes better recommendations than not only a single measure but also the best-seller-based method and that the performance is improved when applying the hybrid approach.

• The Korean Journal of Applied Statistics
• /
• v.28 no.5
• /
• pp.871-883
• /
• 2015

Vector autoregressive (VAR) models in high dimension suffer from noisy estimates, unstable predictions and hard interpretation. Consequently, the sparse vector autoregressive (sVAR) model, which forces many small coefficients in VAR to exactly zero, has been suggested and proven effective for the modeling of high dimensional time series data. This paper studies coupling measures to select non-zero coefficients in sVAR. The basic idea based on the simulation study reveals that removing the effect of other variables greatly improves the performance of coupling measures. sVAR model coefficients are asymmetric; therefore, asymmetric coupling measures such as Granger causality improve computational costs. We propose two asymmetric coupling measures, filtered-cross-correlation and filtered-Granger-causality, based on the filtered residuals series. Our proposed coupling measures are proven adequate for heavy-tailed and high order sVAR models in the simulation study.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.3 no.1
• /
• pp.45-53
• /
• 1995

스파크 점화기관에서 화염전파과정에 관한 연구를 수행하기 위하여는, 초기화염핵 구간에서의 화염의 형성과 발달의 거동을 정확히 파악하여야 한다. 그러므로 화염핵의 형성과 발달에 영향을 미치는 최소 화염핵 크기의 이론적인 계산을 수행하였다. 이론식을 정립하기 위하여 열점화 이론을 이용하였다. 최소 화염핵 크기를 계산하기 위해 열전도 계수, 화염온도, 층류연소속도, 기타 열역학적 상태량 등을 계산하였다. 계산에 의존한 화염핵 크기의 신뢰성을 확인하기 위하여, 점화에너지를 변화시킬 수 있는 점화장치를 사용하여 실기 운전을 통하여 희박연소 한계가 그 때의 화염핵이 성정할 수 있는 영역이라고 가정하여 그 정확도를 확인 하였다.

• Journal of the korean Society of Automotive Engineers
• /
• v.6 no.3
• /
• pp.58-67
• /
• 1984

공간적 화염전파에 대한 실험적인 파악을 통하여 오토엔진에서의 연소과정과 진행을 위한 깊은 통찰을 할 수 있다. 그것을 통하면 매우 희박한 공기, 연료 혼합기의 경우 실린더에서 직접소염 과정을 확인할 수도 있고 hydrocarbon의 불완전 연소와 나타나는 qunching zone간의 관계를 조사할 수도 있다. 광전도 섬유기술(Lichtleit-fasertechnik)을 사용하여 새로 개발된 측정방법을 이용하여 단기통 오토엔진에서 화염면의 공간적인 전파과정과 매우 희박한 공기 연료 혼합기 에서의 quench zone의 출현을 조사하였다. 측정결과들은 공기연료 혼합기가 희박해 질수록 화 염전파 과정이 점점 느려지는 것을 보여준다. 아주 높은 공기 과잉율을 갖는 엔진 운전에서는 화염속도와 연소속도가 매우 급하게 감소한다. 그리하여 화염면은 팽창 단계에서 상대적으로 증가하는 피스톤속도 때문에 더이상 피스톤을 따를 수가 없으며 그로 인해 직접 피스톤상부에 소염대가 형성된다. 그에 의해 배기가스에서의 hydrocarbon 방출의 급격한 증가와 효율이 급격히 감소하는 엔진 운전과 관련이 지어진다.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.11 no.3
• /
• pp.7-12
• /
• 2003

Mixture formation in the cylinder of a lean bum engine has been observed by Laser Induced Fluorescence technique. XeCl laser (308nm) was used to produce a laser sheet. 3-pentanone has been added to iso-octane fuel to produce fluorescence, the intensity of which is proportional to the concentration of the fuel. The laser sheet was introduced through the piston window and the fuel distribution in the vertical plane was observed through a side window. Comparison has been made for the cases of selected fuel injection timing as 0, 360, 405, and 450 CA. For the case of 0 and 360 CA injection, uniform fuel distribution in the combustion chamber has been obtained at the ignition time which is favorable for the high load mode. And the late injection cases, 405 and 450 CA, revealed the stratified formation of rich mixture around the spark plug. That extends the lean misfire limit and reduces cyclic variation in the low load mode.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.2 no.2
• /
• pp.117-123
• /
• 1994

This paper describes the effect of air assisted fuel injection system(AAI) using compressed air to improve the performance of lean combustion engine. AAI is designed to promote fuel atomization and intake flow. In order to investigate the performance of engine with AAl, experiments are conducted varying the engine revolution speed, lean air-fuel ratio and intake manifold pressure. Compared with the original engine, the performance of the engine with MI is improved as the air-fuel mixture becomes leaner or the engine load becomes lower. The descreasing rate of BSFC is propotional to the relative air-fuel ratio and the lean misfire limit extended more than 0.2 relative airfuel ratio.

• Korean Journal of Optics and Photonics
• /
• v.12 no.3
• /
• pp.177-183
• /
• 2001

The KrF laser was employed to study NO fluorescence in lean-bum as well as in soot-bum flames. Blue-shifted NO fluorescence was observed in both of the flames. For both of the flames, the fluorescence intensity of NO and its relative background noise signal were measured with respect to the concentration of seeded NO molecule in the flame and the laser intensity. The results were analyzed qualitatively. Also, NO concentration distribution in the lean-bum flame was qualitatively determined from the intensity of the NO fluorescence. cence.

• Journal of computational fluids engineering
• /
• v.1 no.1
• /
• pp.142-149
• /
• 1996

이차원 노즐을 통하여 저밀도 환경으로 팽창하는 희박류의 분석에 있어서 불연속좌표법과 결합된 유한차분법(finite-difference method coupled with the discrete-ordinate method, FDDO)과 직접모사법(direct-simulation Monte-Carlo method, DSMC)이 비교되었다. FDDO를 이용한 분석에서는 충돌적분모델을 도입하여 간단해진 볼츠만식(Boltzmann equation)이 불연속좌표법을 이용하여 물리적 공간에서는 연속이나 분자속도 공간에서는 불연속좌표로 표시되는 편미분방정식군으로 변환되어 유한차분법에의하여 수치해석 되었다. 직접모사법에서는 분자모델로 가변강구모델(variable hard sphere model, VHS)이, 충돌샘플링모델로는 비시계수법(no time counter method, NTC)이 채택되었다. 전혀 다른 두 가지 방법에 의한 노즐 내부에서의 유체흐름 해석결과는 매우 잘 일치하였으며, 노즐 외부의 plume 영역에서는 FDDO에 의한 해석결과가 직접모사법에 의한 해석결과에 비하여 약간 느린 팽창을 보였다.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.3
• /
• pp.763-770
• /
• 1995

Gas flow characteristics within the cylinder is important factors in impoving lean combustion stability. This paper shows the effects of various flow fields generated by a swirl control valve(SCV) on combustion process in a 4-valve spark ignition engine. An impulse swirl/tumble meter was used to elucidation the steady-state flow characteristics, and a rotating grating type LDV was developed to measure the mean velocity and tunbulence intensity in relation to the crank angle. These methodologies were applied to clarify the correlation between gas flow characteristics and combustion stability at a lean air fuel ratio. An analysis of the correlation revealed the gas flow conditions required to optimize a lean-burn system.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.20 no.11
• /
• pp.3676-3685
• /
• 1996

Enhancement of the ignitability was necessary to realize the lean burn engine. The characteristics of multiple-spark capacitor discharge igniter(MSCDI) usefulness of which for lean burn was examined in constant volume combustion chamber and evaluated in spark ignition engine. Noise of MSCDI for engine was restricted by adoption of low voltage control system. It was found that the adaptability for high engine speed was remarkable. Lean limit in engine with MSCDI was extended 10% than conventional coil ignition system. Also maximum brake thermal efficiency was almost enhanced 1%.