• Journal of KIISE:Software and Applications
• /
• v.32 no.11
• /
• pp.1021-1028
• /
• 2005

The general solution for classification and regression problems can be found by matching and modifying matrices with the information in real world and then these matrices are teaming in neural networks. This paper treats primary space as a real world, and dual space that Primary space matches matrices using kernel. In practical study, there are two kinds of problems, complete system which can get an answer using inverse matrix and ill-posed system or singular system which cannot get an answer directly from inverse of the given matrix. Further more the problems are often given by the latter condition; therefore, it is necessary to find regularization parameter to change ill-posed or singular problems into complete system. This paper compares each performance under both classification and regression problems among GCV, L-Curve, which are well known for getting regularization parameter, and kernel methods. Both GCV and L-Curve have excellent performance to get regularization parameters, and the performances are similar although they show little bit different results from the different condition of problems. However, these methods are two-step solution because both have to calculate the regularization parameters to solve given problems, and then those problems can be applied to other solving methods. Compared with UV and L-Curve, kernel methods are one-step solution which is simultaneously teaming a regularization parameter within the teaming process of pattern weights. This paper also suggests dynamic momentum which is leaning under the limited proportional condition between learning epoch and the performance of given problems to increase performance and precision for regularization. Finally, this paper shows the results that suggested solution can get better or equivalent results compared with GCV and L-Curve through the experiments using Iris data which are used to consider standard data in classification, Gaussian data which are typical data for singular system, and Shaw data which is an one-dimension image restoration problems.

• Journal of the Korean Institute of Gas
• /
• v.16 no.6
• /
• pp.55-66
• /
• 2012

The practical way to minimize the greenhouse gas is to reduce the emission of carbon dioxide. For this reason, CCS(Carbon Capture and Storage) technology, which could reduce carbon dioxide emission, has risen as a realistic alternative in recent years. In addition, the researcher is recently working into ways of applying CCS technologies with deep saline aquifer. In this study, the evaluation model on the feasibility of $CO_2$ sequestration in the deep saline aquifer using ANN(Artificial Neural Network) was developed. In order to develop the efficiency-evaluation model, basic model was created in the deep saline aquifer and sensitivity analysis was performed for the aquifer characteristics by utilizing the commercial simulator of GEM. Based on the sensitivity analysis, the factors and ranges affecting $CO_2$ sequestration in the deep saline aquifer were chosen. The result from ANN training scenario were confirmed $CO_2$ sequestration by solubility trapping and residual trapping mechanism. The result from ANN model evaluation indicated there is the increase of correlation coefficient up to 0.99. It has been confirmed that the developed model can be utilized in feasibility of $CO_2$ sequestration at deep saline aquifer.

• Journal of IKEEE
• /
• v.22 no.1
• /
• pp.162-167
• /
• 2018

In this paper, we propose a method to classify radar signals according to the jamming technique by applying the machine learning to parameter data extracted from received radar signals. In the present army, the radar signal is classified according to the type of threat based on the library of the radar signal parameters mostly built by the preliminary investigation. However, since radar technology is continuously evolving and diversifying, it can not properly classify signals when applying this method to new threats or threat types that do not exist in existing libraries, thus limiting the choice of appropriate jamming techniques. Therefore, it is necessary to classify the signals so that the optimal jamming technique can be selected using only the parameter data of the radar signal that is different from the method using the existing threat library. In this study, we propose a method based on machine learning to cope with new threat signal form. The method classifies the signal corresponding the new jamming method for the new threat signal by learning the classifier composed of the hidden Markov model and the neural network using the existing library data.

• Journal of Digital Convergence
• /
• v.13 no.12
• /
• pp.245-257
• /
• 2015

This study was to develop the predictive model for severity-adjusted mortality of inpatients with multiple chronic conditions and analyse the factors on the variation of hospital standardized mortality ratio(HSMR) to propose the plan to reduce the variation. We collect the data "Korean National Hospital Discharge In-depth Injury Survey" from 2008 to 2010 and select the final 110,700 objects of study who have chronic diseases for principal diagnosis and who are over the age of 30 with more than 2 chronic diseases including principal diagnosis. We designed a severity-adjusted mortality predictive model with using data-mining methods (logistic regression analysis, decision tree and neural network method). In this study, we used the predictive model for severity-adjusted mortality ratio by the decision tree using Elixhauser comorbidity index. As the result of the hospital standardized mortality ratio(HSMR) of inpatients with multiple chronic conditions, there were statistically significant differences in HSMR by the insurance type, bed number of hospital, and the location of hospital. We should find the method based on the result of this study to manage mortality ratio of inpatients with multiple chronic conditions efficiently as the national level. So we should make an effort to increase the quality of medical treatment for inpatients with multiple chronic diseases and to reduce growing medical expenses.

• BMB Reports
• /
• v.53 no.2
• /
• pp.100-105
• /
• 2020

While liver histopathology is heterogeneous in diabetes, the underlying mechanisms remain unclear. We investigated whether glycemic variation resulting from differential diets can induce heterogeneity in diabetic liver and the underlying molecular mechanisms. We generated end-stage non-obese diabetic model rats by subtotal-pancreatectomy in male Sprague-Dawley rats and ad libitum diet for 7 weeks (n = 33). The rats were then divided into three groups, and fed a standard- or a low-protein diet (18 or 6 kcal%, respectively), for another 7 weeks: to maintain hyperglycemia, 11 rats were fed ad libitum (18AL group); to achieve euglycemia, 11 were calorie-restricted (18R group), and 11 were both calorie- and protein-restricted with the low-protein diet (6R group). Overnight-fasted liver samples were collected after the differential diets together with sham-control (18S group), and histology and molecular changes were compared. Hyperglycemic-18AL showed glycogenic hepatopathy (GH) without steatosis, with the highest GSK-3β inactivation because of Akt activation during hyperglycemia; mitochondrial function was not impaired, compared to the 18S group. Euglycemic-18R showed neither GH nor steatosis, with intermediate GSK-3β activation and mitochondrial dysfunction. However, euglycemic-6R showed both GH and steatosis despite the highest GSK-3β activity and no molecular evidence of increased lipogenesis or decreased ApoB expression, where mitochondrial dysfunction was highest among the groups. In conclusion, heterogeneous liver histopathology developed in end-stage non-obese diabetic rats as the glycemic levels varied with differential diets, in which protein content in the diets as well as glycemic levels differentially influenced GSK-3β activity and mitochondrial function in insulin-deficient state.

• The Journal of Korean Medicine
• /
• v.41 no.2
• /
• pp.58-79
• /
• 2020

Objectives: The purpose of this study is to predict the weight loss by applying machine learning using real-world clinical data from overweight and obese adults on weight loss program in 4 Korean Medicine obesity clinics. Methods: From January, 2017 to May, 2019, we collected data from overweight and obese adults (BMI≥23 kg/m2) who registered for a 3-month Gamitaeeumjowi-tang prescription program. Predictive analysis was conducted at the time of three prescriptions, and the expected reduced rate and reduced weight at the next order of prescription were predicted as binary classification (classification benchmark: highest quartile, median, lowest quartile). For the median, further analysis was conducted after using the variable selection method. The data set for each analysis was 25,988 in the first, 6,304 in the second, and 833 in the third. 5-fold cross validation was used to prevent overfitting. Results: Prediction accuracy was increased from 1^st to 2^nd and 3^rd analysis. After selecting the variables based on the median, artificial neural network showed the highest accuracy in 1^st (54.69%), 2^nd (73.52%), and 3^rd (81.88%) prediction analysis based on reduced rate. The prediction performance was additionally confirmed through AUC, Random Forest showed the highest in 1^st (0.640), 2^nd (0.816), and 3^rd (0.939) prediction analysis based on reduced weight. Conclusions: The prediction of weight loss by applying machine learning showed that the accuracy was improved by using the initial weight loss information. There is a possibility that it can be used to screen patients who need intensive intervention when expected weight loss is low.

• Journal of Biomedical Engineering Research
• /
• v.38 no.6
• /
• pp.321-329
• /
• 2017

This paper presents a bone metastasis Detection algorithm on abdominal computed tomography images for early detection using fully convolutional neural networks. The images were taken from patients with various cancers (such as lung cancer, breast cancer, colorectal cancer, etc), and thus the locations of those lesions were varied. To overcome the lack of data, we augmented the data by adjusting the brightness of the images or flipping the images. Before the augmentation, when 70% of the whole data were used in the pre-test, we could obtain the pixel-wise sensitivity of 18.75%, the specificity of 99.97% on the average of test dataset. With the augmentation, we could obtain the sensitivity of 30.65%, the specificity of 99.96%. The increase in sensitivity shows that the augmentation was effective. In the result obtained by using the whole data, the sensitivity of 38.62%, the specificity of 99.94% and the accuracy of 99.81% in the pixel-wise. lesion-wise sensitivity is 88.89% while the false alarm per case is 0.5. The results of this study did not reach the level that could substitute for the clinician. However, it may be helpful for radiologists when it can be used as a screening tool.

• Science of Emotion and Sensibility
• /
• v.13 no.1
• /
• pp.235-242
• /
• 2010

Emotion science is one of the rapidly expanding engineering/scientific disciplines which has a major impact on human society. Such growing interests in emotion science and engineering owe the recent trend that various academic fields are being merged. In this paper we review the recent techniques in the measuring the emotion related elements and applications which include animal model system to investigate the neural network and behaviour, artificial nose/neuronal chip for in-depth understanding of sensing the outer stimuli, metabolic controlling using emotional stimulant such as sounds. In particular, microfabrication techniques made it possible to construct nano/micron scale sensing parts/chips to accommodate the olfactory cells and neuron cells and gave us a new opportunities to investigate the emotion precisely. Recent developments in the measurement techniques will be able to help combine the social sciences and natural sciences, and consequently expand the scope of studies.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.629-635
• /
• 2019

In Deep Learning models derivative is implemented by error back-propagation which enables the model to learn the error and update parameters. It can find the global (or local) optimal points of parameters even in the complex models taking advantage of a huge improvement in computing power. However, deliberately generated data points can 'fool' models and degrade the performance such as prediction accuracy. Not only these adversarial examples reduce the performance but also these examples are not easily detectable with human's eyes. In this work, we propose the method to detect adversarial datasets with random noise addition. We exploit the fact that when random noise is added, prediction accuracy of non-adversarial dataset remains almost unchanged, but that of adversarial dataset changes. We set attack methods (FGSM, Saliency Map) and noise level (0-19 with max pixel value 255) as independent variables and difference of prediction accuracy when noise was added as dependent variable in a simulation experiment. We have succeeded in extracting the threshold that separates non-adversarial and adversarial dataset. We detected the adversarial dataset using this threshold.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.2 s.40
• /
• pp.25-33
• /
• 2006

We propose a method that automatically classifies the images into the object and non-object images. An object image is the image with object(s). An object in an image is defined as a set of regions that lie around center of the image and have significant color distribution against the other surround (or background) regions. We define four measures based on the characteristics of an object to classify the images. The center significance is calculated from the difference in color distribution between the center area and its surrounding region. Second measure is the variance of significantly correlated colors in the image plane. Significantly correlated colors are first defined as the colors of two adjacent pixels that appear more frequently around center of an image rather than at the background of the image. Third one is edge strength at the boundary of candidate for the object. By the way, it is computationally expensive to extract third value because central objects are extracted. So, we define fourth measure which is similar with third measure in characteristic. Fourth one can be calculated more fast but show less accuracy than third one. To classify the images we combine each measure by training the neural network and SYM. We compare classification accuracies of these two classifiers.