• Proceedings of the Korean Statistical Society Conference
• /
• 2005.11a
• /
• pp.91-96
• /
• 2005

다차원층화에서 선형계획법을 이용한 표본배정 방법은 Winkler(1990, 2001), Sitter와 Skinner(1994, 2002)가 제안하였다. 이 방법들은 표본크기가 층 개수보다 크지 않는 경우에 공통적으로 선형계획법을 이용하여 표본배정을 실시하였다. 반복 비율 적합방법(IPF), 일반화 반복 비율 적합(GIFP), SS 방법을 통해 셀 값을 결정하고 선형계획법을 이용하여 표본의 배정확률을 통해 표본배정을 실시한다. 이 3가지 방법들로 표본을 배정하고 평균 및 분산추정량을 비교한다.

• The Korean Journal of Applied Statistics
• /
• v.13 no.1
• /
• pp.197-206
• /
• 2000

For estimating missing cells in contingency table, we suggest an iterative method which extends IPF (Iterative Proportional Fitting) method. The suggested m~thod is not restricted by the number and the location of missing cells, and does not distort the given quasi-independency.

• Communications for Statistical Applications and Methods
• /
• v.17 no.4
• /
• pp.527-540
• /
• 2010

Rule-based classification analysis is widely used for massive datamining because it is easy to understand and its algorithm is uncomplicated. In this classification analysis, majority vote of rules or weighted combination of rules using their supports are frequently used in order to combine rules. We propose a method to combine rules by using the multinomial distribution in this paper. Iterative proportional fitting algorithm is used to estimate the multinomial distribution which maximizes entropy constrained on rules' support. Simulation experiments show that this method can compete with other well known classification models in the case of two similar populations.

• Korean Journal of Organic Agriculture
• /
• v.15 no.3
• /
• pp.319-330
• /
• 2007

This study analyzed physical and chemical characteristics of peat soil to use peat soil as the materials fur pot media and investigated seedling quality of horticultural plants in order to use peat soil as the raw materials fur pot media. The summary of the results is as follows; The chemical characteristics of peat soil, which is main ingredient of pot media are as follows; pH was 4.9, EC was less than $2.0ds{\cdot}m^{-1}$, which interferes the growth of the plant and organic ingredient was 33%. When looking into the germination characteristics of plants according to the mixture of pot media, red pepper showed 54.2% speed of germination and 97% germination rate in peat soil single treatment. Therefore the peatsoil was appropriate for the pot media for red pepper. In case of cucumbers, in the 50 : 50 treatment of main ingredient (peat soil) and auxiliary ingredients (vermiculite, peat moss and perlite) they showed 100% speed of germination and 100% germination rate. Therefore 50 : 50 treatment was appropriate fur the pot media for cucumbers. In case of chinese cabbage, peat soil, perlite and peat moss mixture (50 : 25 : 25) treatment showed the highest speed of germination (77.5%), while the germination rate was a little lower (92.15%) than comparative soil. However, it was appropriate for the pot media for chinese cabbage. In case of watermelon, germination was bad because of the influence of EC when the teat soil ingredient is over 80%. However, in the mixture of peat soil and vermiculite (50:50) treatment, they showed 91.6% speed of germination and 100% germination rate. Therefore it was appropriate for the pot media for watermelon. When looking into the growth of the plants according to the mixture of ingredients, peat soil and perlite (50:50) mixture showed excellent seedling quality for cucumbers, peat soil and perlite (50:50) mixture showed excellent seedling quality and it was proven to be appropriate for the pot media for cucumbers. In case of watermelon, peat soil, peat moss and perlite (80 : 10 : 10) mixture showed excellent seedling quality and it was proven to be appropriate for the pot media for watermelon.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.2
• /
• pp.417-426
• /
• 2024

In general, the implementation of machine learning requires prior knowledge and experience with deep learning models, and substantial computational resources and time are necessary for data processing. As a result, machine learning encounters several limitations when deployed on embedded processors. To address these challenges, this paper introduces a novel approach where a genetic algorithm is applied to the convolution operation within the machine learning process, specifically for performing a selective convolution operation.In the selective convolution operation, the convolution is executed exclusively on pixels identified by a genetic algorithm. This method selects and computes pixels based on a ratio determined by the genetic algorithm, effectively reducing the computational workload by the specified ratio. The paper thoroughly explores the integration of genetic algorithms into machine learning computations, monitoring the fitness of each generation to ascertain if it reaches the target value. This approach is then compared with the computational requirements of existing methods.The learning process involves iteratively training generations to ensure that the fitness adequately converges.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.22 no.4
• /
• pp.361-379
• /
• 2011

This paper aims to propose the most appropriate method for calculating the journal PageRank in a domestic citation database. Korean journals show relatively high journal self-citation ratios and have many outgoing citations to external journals which are not included in the domestic citation database. Because the PageRank algorithm requires recursive calculation to converge, those two characteristics of domestic citation databases must be accounted for in order to measure the citation impact of Korean journals. Therefore, two PageRank calculation methods and four formulas for self-citation adjustment have been examined and tested for KSCD journals. The results of the correlation analysis and regression analysis show that the SCImago Journal Rank formula with the cr2 type self-citation adjustment method seems to be a more appropriate way to measure the relative impact of domestic journals in the Korean Science Citation Database.

• The Korean Journal of Applied Statistics
• /
• v.32 no.5
• /
• pp.763-782
• /
• 2019

This paper proposes a probabilistic document ranking model incorporating term dependencies. Document ranking is a fundamental information retrieval task. The task is to sort documents in a collection according to the relevance to the user query (Qin et al., Information Retrieval Journal, 13, 346-374, 2010). A probabilistic model is a model for computing the conditional probability of the relevance of each document given query. Most of the widely used models assume the term independence because it is challenging to compute the joint probabilities of multiple terms. Words in natural language texts are obviously highly correlated. In this paper, we assume a multinomial distribution model to calculate the relevance probability of a document by considering the dependency structure of words, and propose an information retrieval model to rank a document by estimating the probability with the maximum entropy method. The results of the ranking simulation experiment in various multinomial situations show better retrieval results than a model that assumes the independence of words. The results of document ranking experiments using real-world datasets LETOR OHSUMED also show better retrieval results.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.35 no.5
• /
• pp.641-648
• /
• 2006

In an attempt to get basic data for the utilization of saltwort powder (Salicornia herbaceae L.) as a ingredient in the Sulgidduk. D-optimal design of mixture design showed 14 experimental points including 4 replicates for three independent variables. The three independent variables selected for the experiment were water ($13{\sim}18%$), saltwort powder ($2{\sim}6%$), and sugar ($8{\sim}13%$). The optimum responses variables such as color value. texture, and sensory characteristics were evaluated. The compositional and functional properties of test were measured, and these values were applied to the mathematical models. According to the result of measuring probability of the color value, texture and sensory characteristics were respectively and significance was acknowledged (p<0.05). According to the result of F-test, color values (L, a, b), textural properties (gumminess, chewiness) and sensory characteristics (taste, softness) decided linear model, textural property (hardness) and sensory characteristics (color, smell, moistness, overall acceptance) decided quadratic model. A canonical form and trace plot showed that the influence of each ingredient on the mixture final product. An optimum formulation by numerical and graphical methods were similar. Water, saltwort powder, and sugar were 15.2%, 3.0%, and 9.8% respectively by numerical method, and 15.2%, 3.1%, and 9.7% respectively by graphical method.

• Journal of Korean Ophthalmic Optics Society
• /
• v.17 no.2
• /
• pp.209-213
• /
• 2012

Purpose: To assess the repeatability of the monocular spherical endpoints, a test was performed with four methods which are the retinoscopy, the MPMVA (maximum plus maximum visual acuity) method, the R/G duochrome method, and the crossed cylinder method. Methods: The monocular spherical endpoints was measured by four kinds of method (Retinoscopy, MPMVA method, R/G duochrome method, Crossed cylinder method) on 20 subjects (40 eyes) of average age 23.0 year-old men and women. After a week, retest was performed by same procedure and the test-retest repeatability was assessed by using the Bland-Altman plot analysis. Results: The test-retest mean difference of retinoscopy was the smallest diopters of -0.03 and that of R/G duochrome method was the largest diopters of -0.19. The upper/lower 95% limits of agreement for repeatability was the narrowest in retinoscopy and was the widest in crossed cylinder method. When compared the spherical endpoints of each eye between by retinoscopy and by other three methods, the error rate of ${\pm}0.25D$ in total eyes was 85% in MPMVA method, 80% in R/G duochrome method, and 24% in crossed cylinder method. Conclusions: Test-retest repeatability is the highest in the retinoscopy, and the retinoscopy, the MPMVA method, and R/G duochrome method are suitable for monocular spherical endpoints test.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.21 no.2
• /
• pp.129-139
• /
• 1995

To confirm the usefulness of partial least-squares(PLS) and multiple scattering correction(MSC) method for quantitation of surfactants in [quantitative methods using FT-lR, reconsitituted mixtures of LAS, MES and ELA-9 were tested. Each mixture was dissolved in 50% EtOH, dried, and applied to the KBr cell. From the IR spectra of these mixture, the variance spectrum was obtained. After repeated calibrations for the various regios of this spectrum, we found that 1245-1130cm-1 and 1070-1010cm-1 showed the strong correlation with each component of the sample mixture: all the correlation coefficients were 1.000 and quantitative errors did not exceed 0.32%. From this result, we concluded that PLS method and MSC method are very useful and can be successfully applied to Quality control.