• Journal of the Korean Data and Information Science Society
• /
• v.17 no.1
• /
• pp.95-113
• /
• 2006

The data resulting from a deterministic dynamic system may often appear to be random. However, it is important to distinguish a deterministic and a random processes for statistical analysis. In this paper, we propose a nonparametric test procedure to distinguish a noisy chaos from i.i.d. random process. The proposed procedure can be easily implemented by computer. We notice that the test is very effective to identify a low dimensional chaos process in some cases.

• Journal of the Korean Statistical Society
• /
• v.31 no.1
• /
• pp.1-16
• /
• 2002

Since changepoint identification is important in many data analysis problem, we wish to make inference about the locations of one or more changepoints of the sequence. We consider the Bayesian nonparameteric inference for multiple changepoint problem using a Bayesian segmentation procedure proposed by Yang and Kuo (2000). A mixture of products of Dirichlet process is used as a prior distribution. To decide whether there exists a single change or not, our approach depends on nonparametric Bayesian Schwartz information criterion at each step. We discuss how to choose the precision parameter (total mass parameter) in nonparametric setting and show that the discreteness of the Dirichlet process prior can ha17e a large effect on the nonparametric Bayesian Schwartz information criterion and leads to conclusions that are very different results from reasonable parametric model. One example is proposed to show this effect.

• International Journal of Control, Automation, and Systems
• /
• v.3 no.3
• /
• pp.387-396
• /
• 2005

The Rotational/Translational Actuator (RTAC) benchmark problem considers a fourth-order dynamical system involving the nonlinear interaction of a translational oscillator and an eccentric rotational proof mass. This problem has been posed to investigate the utility of a rotational actuator for stabilizing translational motion. In order to experimentally implement any of the model-based controllers proposed in the literature, the values of model parameters are required which are generally difficult to determine rigorously. In this paper, an approach to the least-squares estimation of the parameters of a system is formulated and practically applied to the RTAC system. On the other hand, this paper shows how to model a nonlinear system as a linear uncertain system via nonparametric system identification, in order to provide the information required for linear robust $H_\infty$ control design. This method is also applied to the RTAC system, which demonstrates severe nonlinearities, due to the coupling from the rotational motion to the translational motion. Experimental results confirm that this approach can effectively condense the whole nonlinearities, uncertainties, and disturbances within the system into a favorable perturbation block.

• Smart Structures and Systems
• /
• v.15 no.2
• /
• pp.409-424
• /
• 2015

Structural nonlinearity is a common phenomenon encountered in engineering structures under severe dynamic loading. It is necessary to localize and identify structural nonlinearities using structural dynamic measurements for damage detection and performance evaluation of structures. However, identification of nonlinear structural systems is a difficult task, especially when proper mathematical models for structural nonlinear behaviors are not available. In prior studies on nonparametric identification of nonlinear structures, the locations of structural nonlinearities are usually assumed known and all structural responses are measured. In this paper, an identification algorithm is proposed for locating and identifying model-free structural nonlinearities and systems using incomplete measurements of structural responses. First, equivalent linear structural systems are established and identified by the extended Kalman filter (EKF). The locations of structural nonlinearities are identified. Then, the model-free structural nonlinear restoring forces are approximated by power series polynomial models. The unscented Kalman filter (UKF) is utilized to identify structural nonlinear restoring forces and structural systems. Both numerical simulation examples and experimental test of a multi-story shear building with a MR damper are used to validate the proposed algorithm.

• 제어로봇시스템학회:학술대회논문집
• /
• 1999.10a
• /
• pp.125-128
• /
• 1999

A linear time-invariant model can be described either by a parametric model or by a nonparametric model. Nonparametric models, for which a priori information is not necessary, are basically the response of the dynamic system such as impulse response model and frequency models. Parametric models, such as transfer function models, can be easily described by a small number of parameters. In this paper aiming to take benefit from both types of models, we will use linear-combination of basis fuctions in an impulse response using a few parameters. We will expand and generalize the Kautz functions as basis functions for dynamical system representations and we will consider estimation problem of transfer functions using Kautz function. And so we will present the influences of poles settings of Kautz function on the identification accuracy.

• Asian Journal of Atmospheric Environment
• /
• v.8 no.3
• /
• pp.115-125
• /
• 2014

In this study, particle composition data for $PM_{10}$ samples were collected every 3 days at Seoul, Korea from August 2006 to November 2007, and were analyzed to provide source identification and apportionment. A total of 164 samples were collected and 21 species (15 inorganic species, 4 ionic species, OC, and EC) were analyzed by particle-induced x-ray emission, ion chromatography, and thermal optical transmittance methods. Positive matrix factorization (PMF) was used to develop source profiles and to estimate their mass contributions. The PMF modeling identified nine sources and the average mass was apportioned to secondary nitrate (9.3%), motor vehicle (16.6%), road salt (5.8%), industry (4.9%), airborne soil (17.2 %), aged sea salt (6.2%), field burning (6.0%), secondary sulfate (16.2%), and road dust (17.7%), respectively. The nonparametric regression (NPR) analysis was used to help identify local source in the vicinity of the sampling area. These results suggest the possible strategy to maintain and manage the ambient air quality of Seoul.

• Communications for Statistical Applications and Methods
• /
• v.15 no.2
• /
• pp.265-271
• /
• 2008

It is commonly believed that diseases of human or economic traits of livestock are caused not by single genes acting alone, but multiple genes interacting with one another. This issue is difficult due to the limitations of parametric-statistic methods of gene effects. So we introduce multifactor-dimensionality reduction(MDR) as a methods for reducing the dimensionality of multilocus information. The MDR method is nonparametric (i. e., no hypothesis about the value of a statistical parameter is made), model free (i. e., it assumes no particular inheritance model) and is directly applicable to case-control studies. Application of the MDR method revealed the best model with an interaction effect between the SNPs, SNP1 and SNP3, while only one main effect of SNP1 was statistically significant for LMA (p < 0.01) under a general linear mixed model.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.111-116
• /
• 2016

의미역은 자연어 문장의 서술어와 관련된 논항의 역할을 설명하는 것으로, 주어진 서술어에 대한 논항인식(Argument Identification) 및 분류(Argument Labeling)의 과정을 거쳐 의미역 결정(Semantic Role Labeling)이 이루어진다. 이를 위해서는 격틀 사전을 이용한 방법이나 말뭉치를 이용한 지도 학습(Supervised Learning) 방법이 주를 이루고 있다. 이때, 격틀 사전 또는 의미역 주석 정보가 부착된 말뭉치를 구축하는 것은 필수적이지만, 이러한 노력을 최소화하기 위해 본 논문에서는 비모수적 베이지안 모델(Nonparametric Bayesian Model)을 기반으로 서술어에 가능한 의미역을 추론하는 비지도 학습(Unsupervised Learning)을 수행한다.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.43 no.5
• /
• pp.145-152
• /
• 2001

농촌유역하천에서 수집된 수질측정자료를 이용하여 주요 오염원을 규명하였다. 분석에 이용된 수질항목은 BOD, COD. EC, pH. TN, TP,$NO_3-N,PO_4-P$ 등으로서 하천 본류내 구간별로 쌍대비교를 통해서 항목별 차이에 대한 유의성 검정을 하였다. 유의성 검정에 이용된 방법은 쌍대비교를 위한 비모수검정법인 Wilcoxon 순위검정법을 이용하였다. 검정결과, 유기성 오염물질의 지표로 이용되는 BOD와 COD가 상승하는 구간은 주로 생활하수가 유입되는 구간으로 나타났다. 인구밀도가 매우 높으나 생활하수가 처리장으로 배제되는 지역의 하천구간에서는 이들 농도변화는 없었다. 유역 내 축산농가가 밀집되어 있는 지역의 하천구간에서도 농도의 변화 또한 존재하지 않았다. 본 연구에서 조사된 유역에서와 같이 축산분뇨를 퇴비화 및 농지환원하는 경우에는 하천수질에 거의 영향을 미치지 않는다는 것을 알 수 있었다. 따라서 농촌지역의 하천 수질을 개선하기 위해서는 생활하수에 대한 처리가 선행되어야 한다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.111-116
• /
• 2016

의미역은 자연어 문장의 서술어와 관련된 논항의 역할을 설명하는 것으로, 주어진 서술어에 대한 논항 인식(Argument Identification) 및 분류(Argument Labeling)의 과정을 거쳐 의미역 결정(Semantic Role Labeling)이 이루어진다. 이를 위해서는 격틀 사전을 이용한 방법이나 말뭉치를 이용한 지도 학습(Supervised Learning) 방법이 주를 이루고 있다. 이때, 격틀 사전 또는 의미역 주석 정보가 부착된 말뭉치를 구축하는 것은 필수적이지만, 이러한 노력을 최소화하기 위해 본 논문에서는 비모수적 베이지안 모델(Nonparametric Bayesian Model)을 기반으로 서술어에 가능한 의미역을 추론하는 비지도 학습(Unsupervised Learning)을 수행한다.