• ETRI Journal
• /
• v.22 no.2
• /
• pp.20-31
• /
• 2000

Much research on natural language processing (NLP), computational linguistics and lexicography has relied and depended on linguistic corpora. In recent years, many organizations around the world have been constructing their own large corporal to achieve corpus representativeness and/or linguistic comprehensiveness. However, there is no reliable guideline as to how large machine readable corpus resources should be compiled to develop practical NLP software and/or complete dictionaries for humans and computational use. In order to shed some new light on this issue, we shall reveal the flaws of several previous researches aiming to predict corpus size, especially those using pure regression or curve-fitting methods. To overcome these flaws, we shall contrive a new mathematical tool: a piecewise curve-fitting algorithm, and next, suggest how to determine the tolerance error of the algorithm for good prediction, using a specific corpus. Finally, we shall illustrate experimentally that the algorithm presented is valid, accurate and very reliable. We are confident that this study can contribute to solving some inherent problems of corpus linguistics, such as corpus predictability, compiling methodology, corpus representativeness and linguistic comprehensiveness.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.05a
• /
• pp.166-169
• /
• 2002

The accuracy in maneuvering target tracking using multiple models is caused by the suitability of each target motion model to be used. The interacting multiple model (IMM) algorithm and the adaptive IMM algorithm require the predefined sub-models and the predetermined acceleration intervals, respectively, in consideration of the properties of maneuvers to construct multiple models. In this paper, to solve these problems intelligently, a genetic algorithm (GA) based-IMM method using fuzzy logic is proposed. In the proposed method, a sub-model is represented as a set of fuzzy rules to model the time-varying variances of the process noises of a new piecewise constant white acceleration model, and the GA is applied to identify this fuzzy model. The proposed method is compared with the AIMM algorithm in simulations.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.49 no.2
• /
• pp.153-160
• /
• 2013

Control allocation is an important part of a system. It implements the function that map the desired command forces from the controller into the commands of the different actuators. In this paper, the authors present an approach for solving constrained control allocation problem in vessel system by using multi-parametric quadratic programming (mp-QP) algorithm. The goal of mp-QP algorithm applied in this study is to compute a solution to minimize a quadratic performance index subject to linear equality and inequality constraints. The solution can be pre-computed off-line in the explicit form of a piecewise linear (PWL) function of the generalized forces and constrains. The efficiency of mp-QP approach is evaluated through a dynamic positioning simulation for a vessel by using four tugboats with constraints about limited pushing forces and found to work well.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2382-2384
• /
• 2002

The accuracy in maneuvering target tracking using multiple models is caused by the suitability of each target motion model to be used. The interacting multiple model (IMM) algorithm and the adaptive IMM (AIMM) algorithm require the predefined sub-models and the predetermined acceleration intervals, respectively, in consideration of the properties of maneuvers in order to construct multiple models. In this paper, to solve these problems intelligently, a genetic algorithm (GA) based-IMM method using fuzzy logic is proposed. In the proposed method, the acceleration input is regarded as an additive noise and a sub-model is represented as a set of fuzzy rules to model the time-varying variances of the process noises of a new piecewise constant white acceleration model. The proposed method is compared with the AIMM algorithm in simulations.

• The KIPS Transactions:PartA
• /
• v.8A no.1
• /
• pp.42-47
• /
• 2001

In this paper, we present a curve fitting algorithm using a ball curve. Our algorithm is recursive method for fitting, which is not a traditional ball function but a continuous ball function. This algorithm consists of two steps. The first step, it is classified the composite corner points to joint points until selected from the given data set. The second step is the curve fitting. The basis function for curve fitting is use to ball function. Also, the weighted least square method, to insert knot, is an efficient method for piecewise ball curve and ball curve segments will be smoothly connected at all composit points. The proposed algorithm will be applied to represent image representation, like fonts, digital image and GIS.

• Korean Journal of Remote Sensing
• /
• v.27 no.2
• /
• pp.141-150
• /
• 2011

Precise image-to-image registration is required to process multi-sensor data together. The purpose of this paper is to develop an algorithm that register between high-resolution optical and SAR images using linear features. As a pre-processing step, initial alignment was fulfilled using manually selected tie points to remove any dislocations caused by scale difference, rotation, and translation of images. Canny edge operator was applied to both images to extract linear features. These features were used to design a cost function that finds matching points based on their similarity. Outliers having larger geometric differences than general matching points were eliminated. The remaining points were used to construct a new transformation model, which was combined the piecewise linear function with the global affine transformation, and applied to increase the accuracy of geometric correction.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.32 no.4
• /
• pp.53-62
• /
• 2009

Lee[15] examined quantity discount contracts between a manufacturer and a retailer in a stochastic, two-period inventory model where quantity discounts are provided based on the previous order size. During the two periods, the retailer faces stochastic (truncated Poisson distributed) demands and he/she places orders to meet the demands. The manufacturer provides for the retailer a price discount for the second period order if its quantity exceeds the first period order quantity. In this paper we extend the above two-period model to a k-period one (where k < 2) and propose a stochastic nonlinear mixed binary integer program for it. In order to make the program tractable, the nonlinear term involving the sum of truncated Poisson cumulative probability function values over a certain range of demand is approximated by an i-interval piecewise linear function. With the value of i selected and fixed, the piecewise linear function is determined using an evolutionary algorithm where its fitness to the original nonlinear term is maximized. The resulting piecewise linear mixed binary integer program is then transformed to a mixed binary integer linear program. With the k-period model developed, we suggest a solution procedure of receding horizon control style to solve n-period (n < k) order decision problems. We implement Lee's two-period model and the proposed k-period model for the use in receding horizon control style to solve n-period order decision problems, and compare between the two models in terms of the pattern of order quantities and the total profits. Our computational study shows that the proposed model is superior to the two-period model with respect to the total profits, and that order quantities from the proposed model have higher fluctuations over periods.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.12 no.2
• /
• pp.5-20
• /
• 1987

This paper addresses sensitivity analysis for a deterministic multi-period production and inventory model. The model assumes a piecewise linear cost structure, but permits backlogging of unsatisfied demand. Our approach to sensitivity analysis here can be divided into two basic steps; (1) to find the optimal production policy through a forward dynamic programming algorithm similar to the backward version of Zangwill [1966] and (2) to apply the penalty network approach by the author [1986] in order to derive sensitivity ranges for various model parameters. Computational aspects are discussed and topics of further research are suggested.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.49 no.8
• /
• pp.374-379
• /
• 2000

The nodal marginal cost and the congestion charge are used as the econimic signals for the electricity price and new invetments in deregulated power systems. In this paper, the nodal marginal cost and the congestion charge are calculated by using the shadow prices resulted from the calculation of Optimal Power Flow(OPF). Linearization of inequality constraints and piecewise linear cost functions make an OPF problem LP-based forms. In order to use the shadow price, the Interior Point(IP) algorithm is applied as a solution technique to the formulation. This paper proposes an algorithm to determine efficients initial points which are guaranteed to be interior points.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.2019-2022
• /
• 2002

This paper presents design of spreading codes for asynchronous DS-CDMA systems. We have been trying to generate maximal-period sequences based on one-dimensional maps with finite bits whose shapes are similar to piecewise linear chaotic maps. We propose an efficient search algorithm finding such mammal-period sequences. This algorithm makes it possible to find many kinds of maximal-period sequences with sufficiently long period for CDMA applications. We also investigate bit error rate(BER) in asynchronous DS-CDMA systems using the maximal-period binary sequences by computer simulations.