• Journal of the Korean Statistical Society
• /
• 제24권1호
• /
• pp.91-109
• /
• 1995

The rate of convergence to a random variable S for an almost certainly convergent series $S_n = \sum^n_{j=1} X_j$ of independent random variables is studied in this paper. More specifically, when $S_n$ converges to S almost certainly, the tail series $T_n = \sum^{\infty}_{j=n} X_j$ is a well-defined sequence of random variable with $T_n \to 0$ a.c. Various sets of conditions are provided so that for a given numerical sequence $0 < b_n = o(1)$, the tail series strong law of large numbers $b^{-1}_n T_n \to 0$ a.c. holds. Moreover, these results are specialized to the case of the weighted i.i.d. random varialbes. Finally, example are provided and an open problem is posed.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1998년도 학술발표대회 논문집 제17권 2호
• /
• pp.369-372
• /
• 1998

본 논문은 어휘독립(Vocabulary-Independent) 환경에서 별도의 훈련과정 없이 인식대상 어휘를 추가 및 변경할 수 있는 가변어휘(Variable Vocabulary) 음성인식에 관한 연구를 다룬다. 가변어휘 인식은 처음에 대용량 음성 데이터베이스(DB)로 음소모델을 훈련하고 인식대상 어휘가 결정되면 발음사전에 의거하여 음소모델을 연결함으로써 별도의 훈련과정 없이 인식대상 어휘를 변경 및 추가할 수 있다. 문맥 종속형(Context-Dependent) 음소 모델인 triphone을 사용하여 인식실험을 하였고, 인식성능의 비교를 위해 어휘종속 모델을 별도로 구성하여 인식실험을 하였다. Unseen triphone 문제와 훈련 DB의 부족으로 인한 모델 파라메터의 신뢰성 저하를 방지하기 위해 state-tying 방법 중 음성학적 지식에 기반을 둔 tree-based clustering(TBC) 기법[1]을 도입하였다. Mel Frequency Cepstrum Coefficient(MFCC)와 대수에너지에 기반을 둔 3 가지 음성특징 벡터를 사용하여 인식 실험을 병행하였고, 연속 확률분포를 가지는 Hidden Markov Model(HMM) 기반의 고립단어 인식시스템을 구현하였다. 인식 실험에는 22 개 부서명 DB[3]를 사용하였다. 실험결과 어휘독립 환경에서 최고 98.4%의 인식률이 얻어졌으며, 어휘종속 환경에서의 인식률 99.7%에 근접한 성능을 보였다.

• 한국항해항만학회지
• /
• 제28권5호
• /
• pp.373-384
• /
• 2004

The purpose of this research is to introduce the concept of "relationship formation factors" in inter-firm relations and to empirically demonstrate that they are effective in improving the relationship results by means of an mediating variable. Therefore, the basic model of this study consists of the independent, mediating, and dependent variables. First, the independent variables, that is, the relationship formation factors, are further classified into three components. transactional, relational, and environmental characteristics. Then, the study empirically examines how each of the three components influences the dependent variable. that is, the business performance(Load Factor), by way of the mediating variables, that is, trust and relationship commitment.

• 한국원자력학회:학술대회논문집
• /
• 한국원자력학회 1997년도 추계학술발표회논문집(1)
• /
• pp.97-102
• /
• 1997

A new method based on the alternating conditional expectation (ACE) algorithm is developed to calculate axial shape index (ASI) for the 3-level excore detector. The ACE algorithm, a type of non-parametric regression algorithms, yields an optimal relationship between a dependent variable and multiple independent variables. In this study, the simple correlation between ASI and excore detector signals is developed using the Younggwang nuclear power plant unit 3 (YGN-3) data without any preprocessing on the relationships between independent variables and dependent variable. The numerical results show that simple correlations exist between the three excore signals and ASI of the core. The accuracy of the new method is much better than those of the current CPC and COLSS algorithms.

• 스마트미디어저널
• /
• 제6권1호
• /
• pp.39-46
• /
• 2017

The internet has become an important part of the many aspects of people's daily lives such as work, study, entertainment and in form of electronic commerce, shopping. Electronic commerce is growing rapidly in Kenya. There are many successful business-to-business, business-to-customer and customer-to-customer online shopping companies in Kenya. As a consequence, competition between vendors is intense and, therefore, mitigating the negative effects of high product uncertainty is necessary requirement as it remains a biggest hindrance for success of the online shopping. The purpose of this research is to investigate how online product description, third party product assurance, customer service and website design mitigate the negative impact of high product uncertainty. A questionnaire with 28 items is designed to collect data from online customers. Using multiple regression analysis, the relationship between dependent variable (product uncertainty) and independent variables such as product description, third party assurance, website design and customer service is tested. The result shows that all the independent variables are negatively correlated with dependent variable, which means that product description, third party assurance, website design and customer service can be used by online vendors to lessen the problem of product uncertainty in online markets.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 1987년도 제29회 수공학연구발표회논문초록집
• /
• pp.223-234
• /
• 1987

A weakly nonlinear solution is presented for the two-dimensional wave kinematics forced by a generic wavemaker of variable-draft. The solution is valid for both piston and hinged wavemakers of variable-draft that may be double articulated. The second-order propagating waves generated by a planar wave board are composed of two components; viz., a Stokes second-order wave and a second-harmonic wave forced by the wavemaker which travels at a different speed. A previously neglected time-independent solution that is required to satisfy a kinematic boundary condition on the wavemaker as well as a mixed boundary condition on the free surface is included for the first time. A component of the time-independent solution is found to accurately estimate the mean return current(correct to second-order) in a closed wave flume. This mean return current is usually estimated from kinematic considerations by a conservation of mass principle.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2022년도 봄 학술논문 발표대회
• /
• pp.208-209
• /
• 2022

This study aims to analyze the risk factors caused by object damage and third-party damage loss in actual bridge construction based on past insurance premium payment data from major domestic insurers for bridge construction projects, and develop a quantitative loss prediction model. For the development of quantitative bridge construction loss model, the dependent variable was selected as the loss ratio, and the independent variable adopted 1) Technical factors: superstructure type, foundation type, construction method, and bridge length 2) Natural hazards: flood anf Typhoon, 3) Project information: total construction duration, total cost and ranking. Among the selected independent variables, superstructure type, construction method, and project period were shown to affect the ratio of bridge construction losses, while superstructure, foundation, flood and ranking were shown to affect the ratio of the third-party losses.

• Journal of Electrical Engineering and Technology
• /
• 제12권2호
• /
• pp.790-802
• /
• 2017

This paper proposes the modeling and control strategy to track the MPPs of hybrid PV and Wind power systems, using a new dual input boost converter. The dual input power conditioning system with an independent MPPT control scheme is introduced with minimum number of circuit elements in order to reduce the switching loss, size and cost of the system. Since the operating conditions for the PV and Wind power systems are very distinct from each other, an efficient and superior control system is required to track the MPPs of both renewable sources with the use of a simply-structured single-ended single-inductor converter. The design of Power-Conditioning System (PCS) and detail control strategy are presented in this paper. To provide independent tracking of MPPs, a variable duty-cycle control strategy is employed for the wind system and a variable frequency strategy is employed for the PV system. Finally, the proposed dual-input converter for hybrid power conditioning system is implemented and the hardware test results are presented. From the hardware experiment, it is concluded that the proposed system successfully tracks the MPPs of both of the renewable power systems independently.

• Journal of Information Technology Applications and Management
• /
• 제18권2호
• /
• pp.61-89
• /
• 2011

Blog is a new global phenomenon, but many research papers about organizational blog have not been shown. In this study the influencing factors of the organizational blog usage are empirically investigated, and several guidelines are suggested to IT professionals who involves the design and implementation of the organizational blog. The research model consists of seven independent variables, one dependent variable, and two moderating variables. The following variables are established as the independent variables; information, interface, service, communication, enjoy, performance expectation, and social influence. Two dimensions such as need and orientation are suggested for the moderating variables, and the actual usage is adopted as a dependent variable. As a result of multiple regression analysis using a stepwise method, the independent variables except for interface and communication affect the actual usage of organizational blogs. The moderating effects for need and orientation are partially supported. The implications of this study are as the followings; (1) The empirical factors affecting the usage of organizational blogs are empirically investigated, (2) The affecting factors vary according to the type of organizational blogs, and (3) Some guidelines are suggested for organizational blog's design.

• 대한간호학회지
• /
• 제14권2호
• /
• pp.84-92
• /
• 1984

This article examined relationships between selected variables, such as demographic background, care, treatment variables, environmental characteristics, and patient's daily behavior and mood change. Relationship were determined between independent variabltherapeutic-rapeutie approach, demographic data, environmental management approach-,and dependent variable-patient's daily behavioral and mood change. 35 patients selected within some criteria in a psychiatric ward, were obserbed during 5 weeks by use of Wyatt's Behavior & Mood Rating Scale ac-cording to the object of the study. At the same time, the frequence of the care and treatment were collected. Criteria for sample selection and independent variables as an influential factor to the patient behavioral change, based on a literature revienw and clinical experiences. Pearson's correlation and multiple regression analysis were used to determine the influfntial factors to the patient behavioral change. Systematic reading (r=.8324), Psychiatrist's individual interview (r=.5764), tranquilizer (r=.3441) and hospitalization processing date (r=.4143) were related with patient's behavioral change. That is these 4 variables can be said to influence to the patient's behavior and mood. A stepwise multiple regression analysis of the effect of the independent varibles of systematic reading, psychintrists individual interview, tranquilizer and hospitalization processing date on the dependent variable, patient's behavioral change was carried out. Systematic reading with on R²of. 69 revealed to be the main influential factor to the patient's behavior and mood change, as the next factor psychiatrist individual interview. A total inclusion of these factors revealed a 73% prediction for the patient's behavior and mood change. But the most influential factor was the interaction of the systematic reading and psychiatrist's individual interview.