• Korean Journal of Cognitive Science
• /
• v.10 no.1
• /
• pp.13-24
• /
• 1999

In this paper, we propose a classifier based on modular networks using an unsupervised learning method. The structure of each module is designed through stochastic analysis of input data and each module classifier data independently. The result of independent classification of each module and a measure of the nearest distance are integrated during the final data classification phase to allow more precise c classification. Computation time is decreased by deleting modules that have been classified to be incorrect during the final classification phase. Using this method. a neural network sharing the best performance was implemented without considering. lots of of variables which can affect the performance of the neural network.

• Journal of Chest Surgery
• /
• v.30 no.9
• /
• pp.908-913
• /
• 1997

From May 1988 to December 1995, 77 patients underwent surgical re ection for primary non-small cell lung cancer at GNUH, and were evaluated clinically. There were 65 males and 12 females(M:P=5.4:1), and the peak incidence of age was 6th decade of life(44.5%). The major symptoms were cough, hemoptysis and chest pain due to anatomical effects of the mass. Histopathologically, squamous cell carcinoma was 81.8%, adenocarcinoma 14.3%, and adenosquamous carcinoma 3.9% . There was no significant difference in survival among three groups. The pneumonectomy was performed in 26 cases(33.8%), lobectomy 30 cases(38.9%), bilobectomy 9 cases(11.7%), and overall resectability was 84.4%. The postoperative official stagings were as follows ; 26 patients of stage I(34%), 14 patients of stage II(18%), 22 patients of stage IIIa(29%), 14 patients of stage IIIb(18%), and one patients of stage IV(1%). In all cases, 3 year survival rate are showed stage 183%, stage II 26%, stage IIIa 17%, and stage IIIb 0%.

• Korean Journal of Plant Taxonomy
• /
• v.37 no.4
• /
• pp.419-430
• /
• 2007

This study was conducted to know the taxonomic features of nuclear ribosomal ITS DNA sequences, ITS1, ITS3 and 5.8S regions, as to nine individuals belonging to four Oxalis species in Korea and an induced species. Sequences of the same regions of sixteen taxa deposited in GenBank were also aligned with those of Korean species as outgroups. The length of ITS sequences aligned in this study is 679 by in total. Evidences, from not only the sequence similarities and divergences but also the phylogenetic and statistical treatments with ITS sequences aligned, were useful for the taxonomy of the genus. The similarity of sequences, among both cauline and acauline taxa, is high as 89% and 95% respectively, but between cauline and acauline taxa, relatively low in the range of 64~69%. The sequence divergences, among both cauline and acauline taxa, is also high as much as 0.36~0.42, but between both cauline and both acauline taxa, low as 0.04~0.06. Two groups between cauline and acauline taxa are paraphyletic, and each group makes a single Glade with a high bootstrap value. The analysis of variance, using ITS sequence aligned, revealed that taxa are significantly different in the level of 0.5%, and O. corymbosa, an induced speices, is also separated from the Korean taxa in the Duncan analysis.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1995.10a
• /
• pp.393-396
• /
• 1995

In this syudy, the researches classifying the artificial and natural flaws in welding parts are performed using the smart pattern recognition technology. For this purpose the smart signal pattern recognition package including the user defined function was developed and the total procedure including the digital signal processing,feature extraction , feature selection and classifier selection is treated by bulk. Specially it is composed with and discussed using the statistical classifier such as the linear disciminant function classifier, the empirical Bayesian classifier. Also, the smart pattern recognition technology is applied to classification problem of natural flaw(i.e multiple classification problem-crack,lack of penetration,lack of fusion,porosity,and slag inclusion, the planar and volumetric flaw classification problem). According to this results, if appropriately learned the neural network classifier is better than ststistical classifier in the classification problem of natural flaw. And it is possible to acquire the recognition rate of 80% above through it is different a little according to domain extracting the feature and the classifier.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.10
• /
• pp.63-70
• /
• 2018

Biomedical signal measurement technology using images has been developed, and researches on respiration signal measurement technology for maintaining life have been continuously carried out. The existing technology measured respiratory signals through a thermal imaging camera that measures heat emitted from a person's body. In addition, research was conducted to measure respiration rate by analyzing human chest movement in real time. However, the image processing using the infrared thermal image may be difficult to detect the respiratory organ due to the external environmental factors (temperature change, noise, etc.), and thus the accuracy of the measurement of the respiration rate is low.In this study, the images were acquired using visible light and infrared thermal camera to enhance the area of the respiratory tract. Then, based on the two images, features of the respiratory tract region are extracted through processes such as face recognition and image matching. The pattern of the respiratory signal is classified through the k-nearest neighbor classifier, which is one of the statistical classification methods. The respiration rate was calculated according to the characteristics of the classified patterns and the possibility of breathing rate measurement was verified by analyzing the measured respiration rate with the actual respiration rate.

• The Korean Journal of Applied Statistics
• /
• v.29 no.5
• /
• pp.859-872
• /
• 2016

Land cover classification is an important tool for preventing natural disasters, collecting environmental information, and monitoring natural resources. Hyperspectral imaging is widely used for this task thanks to sufficient spectral information. However, the curse of dimensionality, spatiotemporal variability, and lack of labeled data make it difficult to classify the land cover correctly. We propose a novel classification framework for land cover classification of hyperspectral data based on convolutional neural networks. The proposed framework naturally incorporates full spectral features with the information from neighboring pixels and has advantages over existing methods that require additional feature extraction or pre-processing steps. Empirical evaluation results show that the proposed framework provides good generalization power with classification accuracies better than (or comparable to) the most advanced existing classifiers.

• The Journal of the Korea Contents Association
• /
• v.9 no.7
• /
• pp.225-232
• /
• 2009

This study conducted the following experiment to examine effects of cardiovascular function on lumbar stabilization exercise(LSE) in floor or swiss ball. This experiment was conducted to compare heart rate, systolic blood pressure, diastolic blood pressure and peripheral vascular oxygen saturation effects by lumbar stabilization exercise in floor or swiss ball with 18 normal adult and it divided 9 subjects. experiment group (1) is applying LSE on floor group and (2) is applying LSE on swiss ball group. Heart rate was measured by portable heart rate manometer, blood pressure was measured by hemodynamometer, and peripheral vascular oxygen concentration was measured using a computerized NURYTEC measuring apparatus analysis. These result lead us to the conclusion that systolic blood pressure and peripheral vascular oxygen concentration were influenced by LSE. but there was not differential effect between each groups. These results suggest that LSE has the capability to improve heart rate, blood pressure, peripheral vascular oxygen concentration. Consequently, LSE would be lead to increment of cardiovascular function.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2009.05a
• /
• pp.1448-1451
• /
• 2009

최근 지구온난화와 기후변화에 관련된 각종 징후들이 여러 분야에서 주요 화두로 자주등장하고 있다. 이들은 주로 평균기온의 상승이나 강우패턴의 변동 등과 같은 기상학적 특성변화를 중심으로 다루어지고 있는데 이를 수문학적 관점에서 유추해 본다면 물 순환과정(hydrological cycle)내 성분별 거동양상의 변화로 해석할 수 있을 것이다. 유역의 특성을 파악하고 발생할 수 있는 수자원의 양적 불균형에 따른 문제점을 탐지하여 그에 대비하기 위해서는 무엇보다도 신속한 정보의 제공이 우선되어야 한다. 또한 이러한 정보를 이용하여 유역의 습윤 및 건조 상황을 모니터링하거나 예측하기 위해서는 즉각적이고 연속적인 정보의 수집이 요구된다. 본 연구에서는 기 수행된 연구결과를 바탕으로 기후학적 물수지 방법에 의하여 1998년부터 2004년까지의 금강유역에 대한 습윤지표를 산정하였다. 그러나 습윤지표가 유역의 습윤 혹은 건조상태를 반영한다고 하나 습윤지표에 익숙하지 않은 사용자의 경우 직관적으로 이 지표만을 이용하여 유역의 상황을 판단하기에는 어려움이 있다. 따라서 본 연구에서는 습윤지표를 통계학적 분포특성에 따라 유역의 습윤 및 건조 상황으로 분류하는 방법을 제안하였으며, 이를 바탕으로 당시 지역적 실제 현상과의 연관성 등을 통하여 가뭄을 평가하는 방법을 제안하고 자 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.04a
• /
• pp.1010-1011
• /
• 2012

적조란 유해조류의 일시적인 대 번식으로 바다를 적색으로 변화시키며 연안 환경 및 바다 생태계에 악영향뿐만 아니라 양식장의 어패류를 집단 폐사 시키는 현상이다. 적조에 의한 양식어업의 피해는 매년 발생하고 있으며 매년 적조방제에 많은 비용을 소비하고 있다. 이 때문에 적조 발생을 미리 예측할 수 있으면 적조에 대한 피해 및 방재 비용을 최소화 시킬 수 있다. 본 논문은 앙상블 학습은 이용한 적조발생 예측 방법을 제안한다. 제안방법은 앙상블 학습의 bagging과 boosting 방법을 이용하여서 적조를 예측의 성능을 향상시킨다. 실험결과 제안방법은 단일 분류기에 비하여서 더 좋은 적조 발생 예측 성능을 보였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.12
• /
• pp.2259-2264
• /
• 2008

Significant genes are defined as genes in which the expression level characterizes a specific experimental condition. Such genes in which the expression levels differ significantly between different groups are highly informative relevant to the studied phenomenon. In this paper, first the system can detect informative genes by similarity scale combination method being proposed in this paper after normalizing data with methods that are the most widely used among several normalization methods proposed the while. And it compare and analyze a performance of each of normalization methods with multi-perceptron neural network layer. The Result classifying in Multi-Perceptron neural network classifier for selected 200 genes using combination of PC(Pearson correlation coefficient) and ED(Euclidean distance coefficient) after Lowess normalization represented the improved classification performance of 98.84%.