• Journal of Korean Institute of Industrial Engineers
• /
• v.32 no.2
• /
• pp.74-81
• /
• 2006

In this paper, we deal with the separation of data by concurrently determined, piecewise nonlinear discriminant functions. Toward the end, we develop a new $l_1$-distance norm error metric and cast the problem as a mixed 0-1 integer and linear programming (MILP) model. Given a finite number of discriminant functions as an input, the proposed model considers the synergy as well as the individual role of the functions involved and implements a simplest nonlinear decision surface that best separates the data on hand. Hence, exploiting powerful MILP solvers, the model efficiently analyzes any given data set for its piecewise nonlinear separability. The classification of four sets of artificial data demonstrates the aforementioned strength of the proposed model. Classification results on five machine learning benchmark databases prove that the data separation via the proposed MILP model is an effective supervised learning methodology that compares quite favorably to well-established learning methodologies.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.11 no.2
• /
• pp.37-50
• /
• 1986

In combining dispersed optimization models, either primal or dual(or both) decomposition method widely used as an organizing device. Interpreting the methods economically, the concepts of price and resource-directive coordination are generally well accepted. Most of deomposition/ integration methods utilize either primal information of dual information, not both, from subsystems, while some authors have developed mixed decomposition approaches employing two master problems dealing primal and dual proposals separately. In this paper a hybrid decomposition method is introduced, where one hybrid master problem utilizes the underlying relationships between primal and dual information from each subsystem. The suggested method is well justified with respect to the flexibility in information flow pattern choice (some prices and other quantities) and to the compatibility of subdivision's optimum to the systemwide optimum, that is often lacking in conventional decomposition methods such as Dantzig-Wolfe's. A numerical example is also presented to illustrate the suggested approach.

• Communications for Statistical Applications and Methods
• /
• v.16 no.6
• /
• pp.971-978
• /
• 2009

Poisson generalized linear mixed models(GLMMs) have been widely used for the analysis of clustered or correlated count data. For the inference marginal likelihood, which is obtained by integrating out random effects is often used. It gives maximum likelihood(ML) estimator, but the integration is usually intractable. In this paper, we propose how to obtain the ML estimator via Laplace approximation based on hierarchical-likelihood (h-likelihood) approach under the Poisson GLMMs. In particular, the h-likelihood avoids the integration itself and gives a statistically efficient procedure for various random-effect models including GLMMs. The proposed method is illustrated using two practical examples and simulation studies.

• Language and Information
• /
• v.5 no.1
• /
• pp.35-53
• /
• 2001

As a subtype of English relative clause constructions, free relative constructions like what John ate in I ate what John ate exhibit complicated syntactic and semantic properties. In particular, the constructions have mixed properties of nominal and verbal: they have the internal syntax of sentence and the external syntax of noun phrase. This paper provides a constraint-based approach to these mixed constructions, and shows that simple constructional constraints are enough to capture their complexities. The paper begins by surveying the properties of the constructions. In discusses two types(Specific and nonspecific) of free relatives, their ,lexical restrictions nominal properties and behavior with respect to extraposition, piped piping and stacking Following these it sketches the basic framework of the HPSG(Head-driven Phrase Structure Grammar) which is of relevance in this paper. As the main part, the paper presents a constraint- based analysis in which tight interactions between grammatical constructions and a rich network of inheritance relations play important roles in accounting for the basic as well as complex properties of the constructions is question.

• Communications for Statistical Applications and Methods
• /
• v.20 no.3
• /
• pp.235-240
• /
• 2013

Generalized linear mixed models(GLMMs) are frequently used for the analysis of longitudinal categorical data when the subject-specific effects is of interest. In GLMMs, the structure of the random effects covariance matrix is important for the estimation of fixed effects and to explain subject and time variations. The estimation of the matrix is not simple because of the high dimension and the positive definiteness; subsequently, we practically use the simple structure of the covariance matrix such as AR(1). However, this strong assumption can result in biased estimates of the fixed effects. In this paper, we introduce Bayesian modeling approaches for the random effects covariance matrix using a modified Cholesky decomposition. The modified Cholesky decomposition approach has been used to explain a heterogenous random effects covariance matrix and the subsequent estimated covariance matrix will be positive definite. We analyze metabolic syndrome data from a Korean Genomic Epidemiology Study using these methods.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.23 no.2
• /
• pp.182-189
• /
• 2009

Nash equilibrium is usually used to investigate a generator's strategic bidding in electricity markets. Some literatures show that the transmission constraints may induce no pure strategy equilibrium and make it hard to find the equilibrium. Using an analytical approach to find a mixed strategy Nash equilibrium in electricity market with transmission constraints, we analyze the influence of transmission capacity on the mixed strategy Nash equilibrium. Finally, a simple numerical example is provided to support the claims of this paper.

• Steel and Composite Structures
• /
• v.27 no.6
• /
• pp.789-801
• /
• 2018

A weak-form formulation of finite annular prism methods (FAPM) based on Reissner's mixed variational theorem (RMVT), is developed for the quasi three-dimensional (3D) static analysis of two-directional functionally graded (FG) circular plates with various boundary conditions and under mechanical loads. The material properties of the circular plate are assumed to obey either a two-directional power-law distribution of the volume fractions of the constituents through the radial-thickness surface or an exponential function distribution varying doubly exponentially through it. These FAPM solutions of the loaded FG circular plates with both simply-supported and clamped edges are in excellent agreement with the solutions obtained using the 3D analytical approach and two-dimensional advanced plate theories available in the literature.

• Proceedings of the KSME Conference
• /
• 2004.04a
• /
• pp.1857-1862
• /
• 2004

EK pumps packed with particles inside capillaries are involved in the mixed electroosmotic flow and pressure driven flow. For analysis in the porous EK pumps, the volume-averaging technique is applied to derive the volume-averaged equations for momentum and electrical potential. By using the volume-averaged equations, analytical solutions for electric potential and velocity distribution due to the mixed electroosmotic and pressure driven flows are obtained. The present analysis is validated by comparison with numerical and experimental results for the case of microchannel EK pumps.

• Proceeding of Spring/Autumn Annual Conference of KHA
• /
• 2001.11a
• /
• pp.33-38
• /
• 2001

The types of the multi-family housing are classified into the independent type, the mixed type and the public type by the approaching method. The characteristics of the independent type are the rear layout type(41.66%), the "ㅋ"type(56.14%), and the 2-room type(54.54%). In the mixed one, they are the front layout type(36.84%), the "ㄱ"type(42.10%), and the 1-room type(49.60%) and in the public one they are the front layout type(51.61%), the "ㅡ"type(41.99%). and the 1-room type(84.11%). It is found that the housing types are gradually tuned into the multi-family housing such as the independent house in the urban area due to changing the major independent type(82.60%) in the early of the 1990s to the public type(74.19%) in the end of the 1990s.

• Journal of Korean Institute of Industrial Engineers
• /
• v.21 no.2
• /
• pp.267-288
• /
• 1995

We investigate a multiclass priority system with a mixed service discipline, and propose a new approach to the analysis of performance measures (mean waiting times) of the system. Customers are preferentially served in the order of priority. The service discipline at each station is either gated or exhaustive discipline. We formulate mean waiting times as functions on the state of the system. We first consider the system at an arbitrary system state to obtain explicit formulae for the mean waiting times, and then derive their steady state values by using the property of Poisson arrivals to see time averages and the generalized Little's formula.