• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.12 no.1
• /
• pp.6-14
• /
• 2012

All types of part-of-speech (POS) tagging errors have been equally treated by existing taggers. However, the errors are not equally important, since some errors affect the performance of subsequent natural language processing seriously while others do not. This paper aims to minimize these serious errors while retaining the overall performance of POS tagging. Two gradient loss functions are proposed to reflect the different types of errors. They are designed to assign a larger cost for serious errors and a smaller cost for minor errors. Through a series of experiments, it is shown that the classifier trained with the proposed loss functions not only reduces serious errors but also achieves slightly higher accuracy than ordinary classifiers.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2009.04a
• /
• pp.496-497
• /
• 2009

In this paper the power flow finite element method (PFFEM) has been used to analyze the vibration of a plate partially covered with a damping sheet. Experiments have been performed to measure the loss factor and frequency response functions of the plate partially covered with the damping sheet. The data for the loss factor has been used as the input data to predict the vibration of the coupled plates with PFFEM. The comparison between the experimental results and the predicted PFFEM results for the frequency response functions has been performed. It showed that PFFEM can be effectively used to predict structural vibration in medium-to-high frequency ranges.

• Korean Chemical Engineering Research
• /
• v.61 no.3
• /
• pp.388-393
• /
• 2023

The analysis of engineering data using neural network based on supervised learning has been utilized in various engineering fields such as optimization of chemical engineering process, concentration prediction of particulate matter pollution, prediction of thermodynamic phase equilibria, and prediction of physical properties for transport phenomena system. The supervised learning requires training data, and the performance of the supervised learning is affected by the composition and the configurations of the given training data. Among the frequently observed engineering data, the data is given in log-scale such as length of DNA, concentration of analytes, etc. In this study, for widely distributed log-scaled training data of virtual 100×100 images, available loss functions were quantitatively evaluated in terms of (i) confusion matrix, (ii) maximum relative error and (iii) mean relative error. As a result, the loss functions of mean-absolute-percentage-error and mean-squared-logarithmic-error were the optimal functions for the log-scaled training data. Furthermore, we figured out that uniformly selected training data lead to the best prediction performance. The optimal loss functions and method for how to compose training data studied in this work would be applied to engineering problems such as evaluating DNA length, analyzing biomolecules, predicting concentration of colloidal suspension.

• Journal of the Korean Statistical Society
• /
• v.30 no.2
• /
• pp.247-263
• /
• 2001

This paper concerns a problem of classification based on training dta. A framework of information geometry is given to elucidate the characteristics of discriminant functions including logistic discrimination and AdaBoost. We discuss a class of loss functions from a unified viewpoint.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.2
• /
• pp.40-44
• /
• 2023

The defect detection based on deep learning requires minimal loss and high accuracy to pinpoint product defects. In this paper, we confirm the loss rate of deep learning training based on disc-shaped artificial scaffold images. It is intended to compare the performance of Cross-Entropy functions used in object detection algorithms. The model was constructed using normal, defective artificial scaffold images and category cross entropy and sparse category cross entropy. The data was repeatedly learned five times using each loss function. The average loss rate, average accuracy, final loss rate, and final accuracy according to the loss function were confirmed.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.451-454
• /
• 2019

In this paper, we propose a verifiable image transformation networks to transform face sketch to photo and vice versa. Face sketch-photo is very popular in computer vision applications. It has been used in some specific official departments such as law enforcement and digital entertainment. There are several existing face sketch-photo synthesizing methods that use feed-forward convolution neural networks; however, it is hard to assure whether the results of the methods are well mapped by depending only on loss values or accuracy results alone. In our approach, we use two Resnet encoder-decoder networks as image transformation networks. One is for sketch-photo and another is for photo-sketch. They depend on each other to verify their output results during training. For example, using photo-sketch transformation networks to verify the photo result of sketch-photo by inputting the result to the photo-sketch transformation networks and find loss between the reversed transformed result with ground-truth sketch. Likely, we can verify the sketch result as well in a reverse way. Our networks contain two loss functions such as sketch-photo loss and photo-sketch loss for the basic transformation stages and the other two-loss functions such as sketch-photo verification loss and photo-sketch verification loss for the verification stages. Our experiment results on CUFS dataset achieve reasonable results compared with the state-of-the-art approaches.

• Korean Journal of Culture and Social Issue
• /
• v.13 no.2
• /
• pp.17-32
• /
• 2007

The word "Loss" means being deprived a person's meaningful property, people or idea by irresistible force, and it is not avoidable in one's life. Especially, the elderly have higher possibility to experience this "Loss" than any other age groups on account of their characteristics. Feeling grief of loss after going through any kind of loss is natural and normal. However, if someone has severe trouble with overcoming this grief, it would affect negatively to his/her psychological or social inelastic. Therefore, we tried to find out which factors consists the "Grief of loss" and how it affects on the elderly's quality of life, especially on the sense of ego-integrity through this study. 97 of the elderly over age 65 participated in the survey and the results as follow; grief of loss is classified into four factors, 'economical loss', 'loss from being parted by death', 'loss of physical functions' and 'loss of relations'. These four factors of "Grief of loss" showed negative correlations with the scores of the sense of ego-integrity factors except 'acceptance of death'. Especially, the 'economical loss' affects on every factor of the sense of ego-integrity negatively except 'acceptance of death' and 'acceptance of aging'. Moreover, 'loss of physical functions' and 'loss of relations' affect negatively on elderly's satisfaction to their lives. On the other hand, the 'loss from being parted by death' of "Grief of loss" and 'acceptance of death' of the sense of ego-integrity showed no statistically significant effect in every process of analysis.

• Korean Journal of Culture and Social Issue
• /
• v.13 no.2
• /
• pp.1-15
• /
• 2007

The purpose of this study is to find how the factors of "Grief of loss" affect on the subjective well-being of elderly('the satisfaction with life', 'the positive affect', 'the negative affect). One hundred of subjects, over age 65, participated in the survey and we could find that the "Grief of loss" was composed of four sub-factors('economical loss', 'loss from being parted by death', 'loss of physical functions', 'loss of relations'). These four sub-factors of "Grief of loss" showed negative correlations with the scores of the satisfaction with life. However, three other sub-factors except 'loss from being parted by death' showed negative correlations with the scores of 'the positive affect'. On the other hand, other three sub-factors except 'economical loss' showed positive correlations with the scores of 'the negative affect'. Especially, the 'economical loss' was affecting on 'the negative affect' and 'the positive affect' significantly. Also, 'loss of relations' was affecting on both 'the negative affect' and 'the satisfaction with life''. Two out of the four sub-factors, 'economical loss' and 'loss from being parted by death', showed no significant effects on the subjective well-being. This results insists us that the emotional supports, especially from their offspring, are truly important for elderly to overcome their Grief of loss('loss of physical functions', 'loss of relations')

• Journal of the Earthquake Engineering Society of Korea
• /
• v.28 no.3
• /
• pp.149-158
• /
• 2024

Piloti-type structures with vertical irregularity are vulnerable to earthquakes due to the soft structure of the first story. Structural characteristics of buildings can significantly affect the seismic loss function, calculated based on seismic fragility, and therefore need to be considered. This study investigated the effects of the number of stories and core locations on the seismic loss function of piloti-type buildings in Korea. Twelve analytical models were developed considering two variations: three stories (4-story, 5-story, and 6-story) and four core locations (center core, x-eccentric core, y-eccentric core, and xy-eccentric core). The interstory drift ratio and peak floor acceleration were assessed through incremental dynamic analysis using 44 earthquake records, and seismic fragility was derived. Seismic loss functions were calculated and compared using the derived seismic fragility and repair cost ratio of each component. The results indicate that the seismic loss function increases with more stories and when the core is eccentrically located in the piloti-type structure model. Therefore, the uncertainty due to the number of stories and core location should be considered when deriving the seismic loss function of piloti-type structures.

• Communications for Statistical Applications and Methods
• /
• v.13 no.1
• /
• pp.89-99
• /
• 2006

This article deals with the problem of estimating parameters of Burr Type XII distribution, on the basis of a general progressive Type II censored sample using Bayesian viewpoints. The maximum likelihood estimator does not admit closed form but explicit sharp lower and upper bounds are provided. Assuming squared error loss and linex loss functions, Bayes estimators of the parameter k, the reliability function, and the failure rate function are obtained in closed form. Finally, a simulation study is also included.