• Korean Journal of Childcare and Education
• /
• v.9 no.2
• /
• pp.117-141
• /
• 2013

The purpose of this study is to investigate the young children's 'more means more' bias and knowledge change process regarding a lever phenomenon, especially the relationship between the weight of an object and the strength of force and between the length of a lever and the strength of force. Subjects, who were presented with the tasks, were eight young children 5 years of age. Major findings were as followings. First, most of the subjects have a 'more means more' bias about the relationship between the weight of an object and the strength of force and between the length of a lever and the strength of force regarding a lever phenomenon. This meant that young children have similar concepts about a lever phenomenon regardless of whether it is right or wrong physically. Second, young children tried to make sure of their knowledge during experiments. They chose the evidence which confirmed their knowledge. But they tried to change their knowledge, when the evidence presented did not correspond to their knowledge. These findings contribute to understanding young children's 'more means more' bias and knowledge change process about a lever phenomenon and can be used in preschool science education programs and curriculums.

• Journal of the Korean Data and Information Science Society
• /
• v.12 no.2
• /
• pp.47-55
• /
• 2001

Usually estimating the means is used for statistical inference. However depending the purpose of survey, sometimes totals will give the better and more meaningful in statistical inference than the means. Here in this study, we dealt with the unemployment population of small areas with using 4 different small area estimation methods: Direct, Synthetic, Composite, Bayes estimation. For all the estimates considered in this study, the average of absolute bias and men square error were obtained in the Monte Carlo Study which was simulated using data from 1998 Economic Active Population Survey in Korea.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.12B
• /
• pp.2292-2297
• /
• 1999

Some observed weather spectra show that nearly 25% of weather spectra are seriously skewed and can not be considered to be symmetric. However, the conventional pulse pair method was derived and has been evaluated under the assumption that the weather spectrum is symmetric and narrow. This means that the conventional pulse pair method may need reevaluation. Therefore, this paper analyzed the bias errors of pulse pair estimates in the skewed spectra. The bias errors of pulse pair mean estimates are more serious comparing with the pulse pair width estimates. In this paper, the poly-pulse pair method is suggested to reduce these bias errors of mean estimates. It was shown that the mean bias errors can be reduced remarkably using the newly suggested poly-pulse pair method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2082-2102
• /
• 2024

Accurate segmentation of magnetic resonance (MR) images is crucial for providing doctors with effective quantitative information for diagnosis. However, the presence of weak boundaries, intensity inhomogeneity, and noise in the images poses challenges for segmentation models to achieve optimal results. While deep learning models can offer relatively accurate results, the scarcity of labeled medical imaging data increases the risk of overfitting. To tackle this issue, this paper proposes a novel fuzzy c-means (FCM) model that integrates a deep learning approach. To address the limited accuracy of traditional FCM models, which employ Euclidean distance as a distance measure, we introduce a measurement function based on the skewed normal distribution. This function enables us to capture more precise information about the distribution of the image. Additionally, we construct a regularization term based on the Kullback-Leibler (KL) divergence of high-confidence deep learning results. This regularization term helps enhance the final segmentation accuracy of the model. Moreover, we incorporate orthogonal basis functions to estimate the bias field and integrate it into the improved FCM method. This integration allows our method to simultaneously segment the image and estimate the bias field. The experimental results on both simulated and real brain MR images demonstrate the robustness of our method, highlighting its superiority over other advanced segmentation algorithms.

• Journal of Korea Water Resources Association
• /
• v.42 no.8
• /
• pp.659-671
• /
• 2009

Fuzzy c-means clustering technique is applied to improve the accuracy of G/R ratio used for rainfall estimation by radar reflectivity. G/R ratio is computed by the ground rainfall records at AWS(Automatic Weather System) sites to the radar estimated rainfall from the reflectivity of Kwangduck Mt. radar station with 100km effective range. G/R ratio is calculated by two methods: the first one uses a single G/R ratio for the entire effective range and the other two different G/R ratio for two regions that is formed by clustering analysis, and absolute relative error and root mean squared error are employed for evaluating the accuracy of radar rainfall estimation from two G/R ratios. As a result, the radar rainfall estimated by two different G/R ratio from clustering analysis is more accurate than that by a single G/R ratio for the entire range.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.4
• /
• pp.435-442
• /
• 2007

To reduce damage from high concentration ozone in the air, we have researched how to predict high concentration ozone before it occurs. High concentration ozone is a rare event and its reaction mechanism has nonlinearities and complexities. In this paper, we have tried to apply and consider as many methods as we could. We clustered the data using the fuzzy c-mean method and took a rejection sampling to fill in the missing and abnormal data. Next, correlations of the input component and output ozone concentration were calculated to transform more correlated components by modified log transformation. Then, we made the prediction models using Dynamic Polynomial Neural Networks. To select the optimal model, we adopted a minimum bias criterion. Finally, to evaluate suggested models, we compared the two models. One model was trained and tested by the transformed data and the other was not. We concluded that the modified transformation effected good to ideal performance In some evaluations. In particular, the data were related to seasonal characteristics or its variation trends.

• Proceedings of the KSRS Conference
• /
• 2009.03a
• /
• pp.254-259
• /
• 2009

In this study, Simplified Method for the Atmospheric Correction (SMAC) radiative transfer model (RTM) used to retrieve surface reflectance from MODIS Top Of Atmosphere (TOA) reflectance (MOD02). SMAC code provides coefficients which were previously yielded by Second Simulation of the Satellite Signal in the Solar Spectrum (6S) for each satellite sensor. We conducted error analysis of SMAC RTM using MOD02 over comparison with MODIS surface reflectance (MOD09) which was provided from 6S. It showed that low accuracy values such as, $R^2$ : 0.6196, Root Means Square Error (RMSE) : 0.00031, bias : - 0.0859. Thus sensitivity analysis of input parameters and coefficients was conducted to searching error sources. Coefficients about $\tau_p$ (average AOD) are more influence than any other coefficients of $\tau_{a550}$ (Aerosol Optical Depth at 550nm) from sensitivity test. Calibrated coefficients of $\tau_p$ from regression analysis were used to surface reflectance which showed that improve accuracy of surface reflectance ($R^2$ : 0.827, RMSE : 0.00672, bias : - 0.000762).

• ETRI Journal
• /
• v.26 no.4
• /
• pp.297-306
• /
• 2004

A plasma is a collection of charged particles and on average is electrically neutral. In fabricating integrated circuits, plasma etching is a key means to transfer a photoresist pattern into an underlayer material. To construct a predictive model of plasma-etching processes, a polynomial neural network (PNN) is applied. This process was characterized by a full factorial experiment, and two attributes modeled are its etch rate and DC bias. According to the number of input variables and type of polynomials to each node, the prediction performance of the PNN was optimized. The various performances of the PNN in diverse environments were compared to three types of statistical regression models and the adaptive network fuzzy inference system (ANFIS). As the demonstrated high-prediction ability in the simulation results shows, the PNN is efficient and much more accurate from the point of view of approximation and prediction abilities.

• English Language & Literature Teaching
• /
• v.11 no.2
• /
• pp.51-66
• /
• 2005

This study investigates the feasibility of incorporating student evaluations of peer review interactions into the course grade for an EFL writing course. The use of such evaluations offers a way to grade the process of writing more directly than using writing portfolios alone. Moreover, evaluating peer review interactions highlights the social aspect of writing, which is valuable in the current post-process climate in writing instruction. The 18 members of a semester-long EFL writing course at a Korean university were trained in peer response for one half of a semester; then performed evaluations of peer review interactions during the second half of the semester as part of their writing course. Student evaluations were examined to reveal whether any bias occurred due to relative age, gender, major, or question type. The results revealed no such biases. Therefore, it is suggested that students are capable of providing fair evaluations of peers, which means the evaluations can be factored into the course grade in order to evaluate the social aspect of the writing process.

• Asian-Australasian Journal of Animal Sciences
• /
• v.15 no.7
• /
• pp.925-931
• /
• 2002

Gibbs sampling algorithms were implemented to the multi-trait threshold animal models with any combinations of multiple binary, ordered categorical, and linear traits and investigate the amount of bias on these models with two kinds of parameterization and algorithms for generating underlying liabilities. Statistical models which included additive genetic and residual effects as random and contemporary group effects as fixed were considered on the models using simulated data. The fully conditional posterior means of heritabilities and genetic (residual) correlations were calculated from 1,000 samples retained every 10th samples after 15,000 samples discarded as "burn-in" period. Under the models considered, several combinations of three traits with binary, multiple ordered categories, and continuous were analyzed. Five replicates were carried out. Estimates for heritabilities and genetic (residual) correlations as the posterior means were unbiased when underlying liabilities for a categorical trait were generated given by underlying liabilities of the other traits and threshold estimates were rescaled. Otherwise, when parameterizing threshold of zero and residual variance of one for binary traits, heritability estimates were inflated 7-10% upward. Genetic correlation estimates were biased upward if positively correlated and downward if negatively correlated when underling liabilities were generated without accounting for correlated traits on prior information. Residual correlation estimates were, consequently, much biased downward if positively correlated and upward if negatively correlated in that case. The more categorical trait had categories, the better mixing rate was shown.