• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11c
• /
• pp.1367-1370
• /
• 2003

본 논문에서는 인덱스가 존재하지 않는 두 개의 입력 릴레이션에 대해서도 최적의 조인 연산을 수행할 수 있는 공간 해쉬 조인 알고리즘을 제안한다. 인덱스가 존재하지 않는 릴레이션의 처리에 사용하는 기존의 공간 해쉬 조인(SHJ: Spatial Hash Join)과 Scalable Sweeping-Rased Spatial Join(SSSJ) 알고리즘을 결합하여 SHJ 알고리즘의 단점으로 지적되고 있는 편향된(skewed) 데이터에 대한 조인 연산의 성능저하 문제를 개선한 수 있는 Spatial Hash Strip Join(SHSJ) 알고리즘을 제안한다. SHJ에서 편향된 데이터의 경우 해쉬 버킷의 오버플로우 처리를 위해 버킷 재분할 방법을 사용하고 있는데 반하여 본 논문에서 제안한 SHSJ 알괴리즘에서는 버킷의 재분할 처리 대신에 버킷에 데이터를 삽입하고, 조인 연산과정에서 오버플로우가 발생한 버킷에 대하여 SSSJ 알고리즘을 사용함으로써 편향된 입력 릴레이션의 처리 성능을 제고시킬 수 있도록 한다.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.47 no.1
• /
• pp.1-8
• /
• 2019

An experimental study is performed to observe the characteristics of the thrust vectoring control (TVC) of the supersonic jet using proportional control valves. It is observed that three different TVC characteristics exist as the nozzle pressure ratio varies. Strong hysteresis phenomena are also observed during the valve control for a certain range of the nozzle pressure ratio. It is also noticed that the secondary chamber pressure is one of the influencing parameters for the TVC. Therefore, a control algorithm utilizing the secondary chamber pressure coefficient as a predictor is applied to achieve the stable TVC avoiding the hysteresis. Consequently, the stable TVC with the maximum deflection angle of about 20-degree has been realized using the proportional control valves.

• Geophysics and Geophysical Exploration
• /
• v.13 no.1
• /
• pp.118-127
• /
• 2010

Resistivity logging instruments are designed to measure the electrical resistivity of a formation, and this can be directly interpreted to provide a water-saturation profile. However, resistivity logs are sensitive to borehole and shoulder-bed effects, which often result in misinterpretation of the results. These effects are emphasised more in the presence of tool eccentricity. For precise interpretation of short- and long-normal logging measurements in the presence of tool eccentricity, we simulate and analyse eccentricity effects by combining the use of a Fourier series expansion in a new system of coordinates with a 2D goal-oriented high-order self-adaptive hp finite-element refinement strategy, where h denotes the element size and p the polynomial order of approximation within each element. The algorithm automatically performs local mesh refinement to construct an optimal grid for the problem under consideration. In addition, the proper combination of h and p refinements produces highly accurate simulations even in the presence of high electrical resistivity contrasts. Numerical results demonstrate that our algorithm provides highly accurate and reliable simulation results. Eccentricity effects are more noticeable when the borehole is large or resistive, or when the formation is highly conductive.

• The Korean Journal of Applied Statistics
• /
• v.17 no.3
• /
• pp.459-473
• /
• 2004

It has well known that an exhaustive search algorithm suggested by Breiman et. a1.(1984) has a trend to select the variable having relatively many possible splits as an splitting rule. We propose an algorithm to overcome this variable selection bias problem and then construct unbiased regression trees based on the algorithm. The proposed algorithm runs two steps of selecting a split variable and determining a split rule for binary split based on the split variable. Simulation studies were performed to compare the proposed algorithm with Breiman et a1.(1984)'s CART(Classification and Regression Tree) in terms of degree of variable selection bias, variable selection power, and MSE(Mean Squared Error). Also, we illustrate the proposed algorithm with real data sets.

• Journal of KIISE
• /
• v.42 no.9
• /
• pp.1124-1132
• /
• 2015

In order to train a supervised machine learning algorithm, it is necessary to have non-biased labels and a sufficient amount of training data. However, it is difficult to collect the required non-biased labels and a sufficient amount of training data to develop an automatic English Composition scoring system. In addition, an English writing assessment is carried out using a multi-faceted evaluation of the overall level of the answer. Therefore, it is difficult to choose an appropriate machine learning algorithm for such work. In this paper, we show that it is possible to alleviate these problems through ensemble learning. The results of the experiment indicate that the ensemble technique exhibited an overall performance that was better than that of other algorithms.

• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.177-190
• /
• 2021

Random forests is a popular method that improves the instability and accuracy of decision trees by ensembles. In contrast to increasing the accuracy, the ease of interpretation is sacrificed; hence, to compensate for this, variable importance is provided. The variable importance indicates which variable plays a role more importantly in constructing the random forests. However, when a predictor is correlated with other predictors, the variable importance of the existing importance algorithm may be distorted. The downward bias of correlated predictors may reduce the importance of truly important predictors. We propose a new algorithm remedying the downward bias of correlated predictors. The performance of the proposed algorithm is demonstrated by the simulated data and illustrated by the real data.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.493-495
• /
• 2015

PDR은 일반적으로 IMU센서로 부터의 가속도와 각속도를 측정하여 보행자의 위치를 추적하는 시스템이다. IMU센서로부터 측정된 가속도와 각속도 값은 센서를 기준으로 하기 때문에 보행자가 인지하는 고정 좌표계와는 차이가 있다. 이를 해결하기 위해 회전행렬을 사용하며 이후 계속해서 측정되는 각속도를 통해 회전행렬을 업데이트 한다. 업데이트된 회전행렬을 통해 좌표계를 환산하고 환산된 좌표계의 가속도 값으로부터 보행자는 고정좌표계 기준으로 위치 추적이 가능하다. 하지만 회전행렬을 업데이트 하는 과정에서 센서의 세 축이 이상적으로 수직이 아니라면 업데이트 과정에서 각속도의 오차가 누적되고 이는 좌표계를 환산에 영향을 끼쳐 위치 및 속도 추적 정확성을 낮춘다. 물리적인 Bias가 PDR 시스템에 누적오차를 발생시킨다. 이에 제안하는 센서 축 편향 보정 알고리즘은 IMU 센서의 물리적 축 오차를 보정해주어 더 정확한 위치 추적을 가능하게 한다. 또한 Matlab을 통해 데이터를 분석하고 알고리즘의 필요성을 보인다.

• Journal of Digital Convergence
• /
• v.14 no.1
• /
• pp.171-179
• /
• 2016

The energy consumption problem in wireless sensor networks is investigated. The problem is to expend as little energy as possible receiving and transmitting data, because of constrained battery. In this paper, in order to extend the lifetime of the network, we proposed a location-based clustering algorithm for wireless sensor network with skewed-topology. The proposed algorithm is to deploy multiple child nodes at the sink to avoid bottleneck near the sink and to save energy. Proposed algorithm can reduce control traffic overhead by creating a dynamic cluster. We have evaluated the performance of our clustering algorithm through an analysis and a simulation. We compare our algorithm's performance to the best known centralized algorithm, and demonstrate that it achieves a good performance in terms of the life time.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.3
• /
• pp.11-17
• /
• 2020

In this paper, we propose practical results for cache replacement policy by measuring cache hit and search time for each replacement algorithm through cache simulation. Thus, the structure of each cache memory and the four types of alternative policies of FIFO, LFU, LRU and Random were implemented in software to analyze the characteristics of each technique. The paper experiment showed that the LRU algorithm showed hit rate and search time of 36.044% and 577.936ns in uniform distribution, 45.636% and 504.692ns in deflection distribution, while the FIFO algorithm showed similar performance to the LRU algorithm at 36.078% and 554.772ns in even distribution and 45.662% and 489.574ns in bias distribution. Then LFU followed, Random algorithm was measured at 30.042% and 622.866ns at even distribution, 36.36% at deflection distribution and 553.878ns at lowest performance. The LRU replacement method commonly used in cache memory has the complexity of implementation, but it is the most efficient alternative to conventional alternative algorithms, indicating that it is a reasonable alternative method considering the reference information of data.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.31-33
• /
• 2005

GPS및 PDA의 발달로 인해서 위치 기반 서비스(LBS), 차량항법장치(CNS), 지리정보시스템(GIS)등 공간 데이터를 다루는 응용프로그램들이 급속하게 보급되었다. 이러한 응용프로그램은 높이 균등 색인 기법을 사용하여 원하는 데이터에 대한 색인을 제공하였다. 그러나 모든 공간 객체는 서로 상이한 접근 빈도를 가지고 있음에도 불구하고 기존의 공간색인 기법은 접근 빈도를 고려하지 못하는 단점을 가지고 있었다. 또한 기존의 빈도수만을 고려한 공간 객체의 색인 방법은 접근 빈도에 따른 편향성(skewed)은 제공하지만 공간 객체에 대한 지역성을 반영하지 못한다. 본 논문에서는 밀집되어 있는 공간 객체의 접근 빈도를 반영해서 편향된 색인 트리를 생성하는 기법을 제안한다. 이형 클러스터링으로 분포되어 있는 전체 영역에 대해서 Zahn의 클러스터링 알고리즘을 변형시켜서 다단계 세부영역을 구분한다. 이렇게 구간된 세부영역에 대해서 거리적 인접성과 접근 빈도수의 합을 이용해서 색인 트리를 생성한다. 다단계로 구성된 전체영역에 대해서 하향식 방식으로 편향된 색인 트리를 생성함으로써, 접근 빈도가 높은 공간 객체에 대해서 빠른 탐색이 가능하게 한다.