• Proceedings of the IEEK Conference
• /
• 2000.11d
• /
• pp.167-170
• /
• 2000

In this paper, we present a fast algorithm for the motion estimation using the efficient selection of an initial search position. In the method, we select the initial search position using the motion vector from the subsmpled images, the predicted motion vector from the neighbor blocks, and the (0,0) motion vector. While searching the candidate blocks, we use the spiral search pattern with the successive elimination algorithm(SEA) and the partial distortion elimination(PDE). The experiment results show that the complexity of the proposed algorithm is about 2∼3 times faster than the three-step search(TSS) with the PSNR loss of just 0.05［dB］∼0.1［dB］ than the full search algorithm PSNR. The search complexity can be reduced with quite a few PSNR loss by controling the number of the depth in the spiral search pattern.

• Journal of the Korean Statistical Society
• /
• v.19 no.1
• /
• pp.24-44
• /
• 1990

The purpose of this paper is to extend the idea of Tamhane and Bechhofer (1977, 1979) concerning the normal means problem to some general class of distributions. The key idea in Tamhane and Bechhofer is the derivation of the computable lower bounds on the probability of a correct selection. To derive such lower bounds, they used the specific covariance structure of a multivariate normal distribution. It is shown that such lower bounds can be obtained for a class of stochastically increasing distributions under certain conditions, which is sufficiently general so as to include the normal means problem as a special application. As an application of the general theory to the scale parameters problem, a two-stage elimination type procedure for selecting the population associated with the smallest variance from among several normal populations is proposed. The design constants are tabulated and the relative efficiencies are computed.

• International Journal of Internet, Broadcasting and Communication
• /
• v.9 no.1
• /
• pp.72-80
• /
• 2017

In this work, an efficient transmit antenna selection approach for the quadrature spatial modulation (QSM) systems is proposed. The conventional Euclidean distance antenna selection (EDAS)-based schemes in QSM have too high computational complexity for practical use. The proposed antenna selection algorithm is based on approximation of the EDAS decision metric employed for QSM. The elimination of imaginary parts in the decision metric enables decoupling of the approximated decision metric, which enormously reduces the complexity. The proposed method is also evaluated via simulations in terms of symbol error rate (SER) performance and compared with the conventional EDAS methods in QSM systems.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2004.04a
• /
• pp.87-94
• /
• 2004

A number of approximate techniques have been developed to calculate the eigenvalues in a reduced manner. These schemes approximate the lower eigenvalues that represent the global behavior of the structures. In general, sequential elimination has been widely used with reliability. But it takes excessively large amount of time to construct a reduced system. The present study proposes two-level condensation scheme(TLCS). In the first step, the candidate elements are selected by element-level energy estimation. In the second step, master degrees of freedom are selected by sequential elimination from the candidate degrees of freedom linked to the selected elements in the first step. Numerical examples demonstrate that the proposed method saves computational cost effectively and provides a reduced system which predicts the accurate eigenvalues of global system.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.2
• /
• pp.91-97
• /
• 2019

In automatic map generalization, the formalization of cartographic principles is important. This study proposes and evaluates the selection method for road network generalization that analyzes existing maps using reverse engineering and formalizes the selection rules for the road network. Existing maps with a 1:5,000 scale and a 1:25,000 scale are compared, and the criteria for selection of the road network data and the relative importance of each network object are determined and analyzed using $T{\ddot{o}}pfer^{\prime}s$ Radical Law as well as the logistic regression model. The selection model derived from the analysis result is applied to the test data, and road network data for the 1:25,000 scale map are generated from the digital topographic map on a 1:5,000 scale. The selected road network is compared with the existing road network data on the 1:25,000 scale for a qualitative and quantitative evaluation. The result indicates that more than 80% of road objects are matched to existing data.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2000.11a
• /
• pp.469-477
• /
• 2000

본 연구에서는 부도예측용 인공신경망 모형의 입력노드를 선정하기 위한 방법론으로 연결강도판별분석(Link Weight Discrimination Analysis)에 의한 약체뉴론제거법(Weak-Linked Neuron Elimination)과 강체뉴론선택법 (Strong-Linked Neurons Selection)을 제안한다. 연결강도판별분석이란 적절한 학습이 끝난 인공신경망 모형에서 입력노드와 연결되는 가중치의 합에 대한 절대값인 연결강도 판별식(Link Weight Discrimination)에 의해 해당 입력노 드가 출력노드에 미치는 영향정도를 분석하는 것이다. 한편 강체연결뉴론선택법은 선처리를 통해 얻어진 학습된 인공신경망의 입력노드 가운데서 연결강도판별식이 큰 뉴론만을 본처리의 입력노드로 선정하는 것인데 비해 약체연결뉴론제거법은 연결강도판별식이 일정 값 즉, 연결강도 판별임계치(Link Weight Discrimination Cut off Value) 보다 낮은 입력노드를 제외하고 나머지 입력노드만을 본처리의 입력노드로 선정하는 것이다. 본 연구에서는 강체연결뉴론선택법과 약체연결뉴론제거법을 각각 정형적인 방법론으로 정립하고 이 방법론에 의해 부도예측용 인공신경망을 구축하여 각각의 모형을 의사결정트리에 의해 선정된 인공신경망 모형 및 선처리 과정을 거치지 않은 인공신경망 모형과 성능을 비교, 분석하여 본 연구에서 제안한 방법론의 타당성을 제시하였다.

• Journal of Sensor Science and Technology
• /
• v.25 no.1
• /
• pp.13-19
• /
• 2016

Nowadays, most major automobile manufacturers are very interested, and actively involved, in developing driver alcohol detection system for safety (DADSS) that serves to prevent driving under the influence. DADSS measures the blood alcohol concentration (BAC) from the driver's breath and limits the ignition of the engine of the vehicle if the BAC exceeds the reference value. In this study, to optimize the sensor array of the DADSS, we selected sensors by using three different methods, configured the sensor arrays, and then compared their performance. The Wilks' lambda, stepwise elimination and filter method (using a principal component) were used as the sensor selection methods [2,3]. We compared the performance of the arrays, by using the selectivity and sensitivity as criteria, and Sammon mapping for the analysis of the cluster type of each gas. The sensor array configured by using the stepwise elimination method exhibited the highest sensitivity and selectivity and yielded the best visual result after Sammon mapping.

• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.82-89
• /
• 2023

Various machine-learning models may yield high predictive power for massive time series for time series prediction. However, these models are prone to instability in terms of computational cost because of the high dimensionality of the feature space and nonoptimized hyperparameter settings. Considering the potential risk that model training with a high-dimensional feature set can be time-consuming, we evaluate a feature-importance-based feature selection method to derive a tradeoff between predictive power and computational cost for time series prediction. We used two machine learning techniques for performance evaluation to generate prediction models from a retail sales dataset. First, we ranked the features using impurity- and Local Interpretable Model-agnostic Explanations (LIME) -based feature importance measures in the prediction models. Then, the recursive feature elimination method was applied to eliminate unimportant features sequentially. Consequently, we obtained a subset of features that could lead to reduced model training time while preserving acceptable model performance.

• The Korean Journal of Applied Statistics
• /
• v.18 no.2
• /
• pp.435-443
• /
• 2005

Coefficients of determination in logistic regression analysis are defined as various statistics, and their values are relatively smaller than those for linear regression model. These coefficients of determination are not generally used to evaluate and diagnose logistic regression model. Liao and McGee (2003) proposed two adjusted coefficients of determination which are robust at the addition of inappropriate predictors and the variation of sample size. In this work, these adjusted coefficients of determination are applied to variable selection method for logistic regression model and compared with results of other methods such as the forward selection, backward elimination, stepwise selection, and AIC statistic.

• Geomechanics and Engineering
• /
• v.25 no.1
• /
• pp.1-16
• /
• 2021

Machine learning models have been widely used for landslide susceptibility assessment (LSA) in recent years. The large number of inputs or conditioning factors for these models, however, can reduce the computation efficiency and increase the difficulty in collecting data. Feature selection is a good tool to address this problem by selecting the most important features among all factors to reduce the size of the input variables. However, two important questions need to be solved: (1) how do feature selection methods affect the performance of machine learning models? and (2) which feature selection method is the most suitable for a given machine learning model? This paper aims to address these two questions by comparing the predictive performance of 13 feature selection-based machine learning (FS-ML) models and 5 ordinary machine learning models on LSA. First, five commonly used machine learning models (i.e., logistic regression, support vector machine, artificial neural network, Gaussian process and random forest) and six typical feature selection methods in the literature are adopted to constitute the proposed models. Then, fifteen conditioning factors are chosen as input variables and 1,017 landslides are used as recorded data. Next, feature selection methods are used to obtain the importance of the conditioning factors to create feature subsets, based on which 13 FS-ML models are constructed. For each of the machine learning models, a best optimized FS-ML model is selected according to the area under curve value. Finally, five optimal FS-ML models are obtained and applied to the LSA of the studied area. The predictive abilities of the FS-ML models on LSA are verified and compared through the receive operating characteristic curve and statistical indicators such as sensitivity, specificity and accuracy. The results showed that different feature selection methods have different effects on the performance of LSA machine learning models. FS-ML models generally outperform the ordinary machine learning models. The best FS-ML model is the recursive feature elimination (RFE) optimized RF, and RFE is an optimal method for feature selection.