• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.395-396
• /
• 2019

최근 컴퓨터 하드웨어, 소프트웨어의 급속한 발전으로 상용화되면서 생체 인식 기술은 몇 년 전부터 점차 넓은 시장을 형성하고 있다. 본 논문에서는 얼굴 인식을 위하여 학습 데이터구성과 특징데이터에 따른 인식 정도를 파악하고 효과적인 방법으로 학습할 수 있는 방법을 제안하고자 한다. 실험결과, 원영상 그대로 인식하는 것 보다 특징 데이터를 구성하여 학습하는 것이 효율적임을 알 수 있다.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.1277-1280
• /
• 1998

We propose a structural learning method of MLP classifiers for a given application using PfSGA (parameter-free species genetic algorithm), which is a combining of species genetic algorithm(SGA) and parameter-free genetic algorithm(PfGA). experimental results show that PfSGA can reduce the learing time of SGA and has no influence of parameter values on structural learning. And we also convince that PfSGA is more efficient than the other methods in the aspect of misclassification ratio, learning rate, and complexity of MLP structure.

• Journal of Digital Contents Society
• /
• v.8 no.3
• /
• pp.355-363
• /
• 2007

One of major factors for learning achievement is the student's learning preference according to his character type. In course of learning, if a student studies e-learning contents opposed to his preference, then he would be under stress and his blood pressure and heart beat be changed. For measuring unwillingness, we used spectral components in frequency domain known as stress measure. For 13 children attending kindergarten we examined S(sensing)/ N(intuition) of MBTI and presented same learning contents during 10 minutes. During learning we gathered ECG signals, changed into HRV(heart rate variability), transformed time-varying HRV signal into spectral density in frequency domain. And then, we divided it into three areas of low(LF), middle(MF), and high-frequency(HF) and calculated stress measures by rates of those frequency area. We compared estimated stress measures of S group with them of N group whether students in different group preferred different contents or not. Experimental shows that students according to MBTI type prefer different contents.

• Journal of the Korean Home Economics Association
• /
• v.28 no.2
• /
• pp.91-106
• /
• 1990

This study focuses on consumers' advertising discriminatory competencies and the influencing facors. So the objects of this study are as follows : 1) to identify the overall level of advertising discriminatory competencies. 2) to examine if consumer attitude variables have significant effects on the ads. discriminatory competenceis. 3) to examine if the frequencies of contacting advertising variable have significant effects on the ads. discriminatory competencies. 4) to examine if socio-economics variables-age, educational level, monthly family income, occupational status-have significant effects on the ads. discriminatory competencies. 5) to find out the independent influence of variables related to the ads. discriminatory competencies. For this purpose, a survey was conducted using questionaires and advertisement papers. The data used in this study included 194 Homemakers living in Seoul. The ads. used in this study included ads. of diary products like foods, drinks, medicine, cosmetic, detergent in TV, radio, newspaper and magagine. Statistics were Frequency Distribution, Mean, percentile, ANOVA, Scheff -test, Pearsons' Correlation, Multiple Regrassion Analysis. Major findings were as follows : First, in 26 items(70%) of 37 items measured consumers' ads. discriminatory competencies, the rate of right answer was below 50%, so over the half of consumers were misleaded by the deceptive ads. Second, consumers' ads. discriminatory competencies differed significantly according to consumer attitude variables but no according to the frequency of contacting advertising. Third, according to socio-demographic variables-age, educational level, monthly family income, occupational status-advertising discriminatory competencies differed significantly. In group of lower age, higher educational level, higher income and professional occupation status, the level of ads. discriminatory competencies were high. Forth, the most influencing variabel on ads. disciriminatory competencies were eudcational level and in turn general attitude toward ads., attitude toward consumerism. This three variables explain 22.9% of dependent variable's variance. From these findings, the following suggestions are made, First, the consumer education offering informations and learing practical ads. discriminatory competencies should be conducted for all consumers wheather they are educated or not. Also the education to improve the consumer attitude must be. Second, considering misleading level, the business must make the regulatory standards and reinforce the regulation voluntarily, and by enforcing the regulation of ads. and deciding more diverse, objective and exact standards, the government should keep the consumer's right to know.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.763-768
• /
• 2023

In this paper, the existing statistical method (ARIMA) and machine learning method (Informer) were developed and compared to predict the distribution volume of pharmaceuticals. It was found that a machine learning-based model is advantageous for daily data prediction, and it is effective to use ARIMA for monthly prediction and switch to Informer as the data increases. The prediction error rate (RMSE) was reduced by 26.6% compared to the previous method, and the prediction accuracy was improved by 13%, resulting in a result of 86.2%. Through this thesis, we find that there is an advantage of obtaining the best results by ensembleing statistical methods and machine learning methods. In addition, machine learning-based AI models can derive the best results through deep learning operations even in irregular situations, and after commercialization, performance is expected to improve as the amount of data increases.

• Journal of Korean Society of Water and Wastewater
• /
• v.37 no.5
• /
• pp.271-279
• /
• 2023

Fouling is an inevitable problem in membrane water treatment plant. It can be measured by trans-membrane pressure (TMP) in the constant flux operation, and chemical cleaning is carried out when TMP reaches a critical value. An early fouilng alarm is defined as warning the critical TMP value appearance in advance. The alarming method was developed using one of machine learning algorithms, decision tree, and applied to a ceramic microfiltration (MF) pilot plant. First, the decision tree model that classifies the normal/abnormal state of the filtration cycle of the ceramic MF pilot plant was developed and it was then used to make the early fouling alarm method. The accuracy of the classification model was up to 96.2% and the time for the early warning was when abnormal cycles occurred three times in a row. The early fouling alram can expect reaching a limit TMP in advance (e.g., 15-174 hours). By adopting TMP increasing rate and backwash efficiency as machine learning variables, the model accuracy and the reliability of the early fouling alarm method were increased, respectively.