• 한국신뢰성학회지:신뢰성응용연구
• /
• 제17권3호
• /
• pp.188-196
• /
• 2017

Purpose: The purpose of this research is to determine optimal replacement age using non-informative prior information and Bayesian method. Methods: We propose a novel approach using Bayesian method to determine the optimal replacement age in block replacement policy by defining the prior probability with data on failure time and repair time. The Marcov Chain Monte Carlo simulation is used to investigate the asymptotic distribution of posterior parameters. Results: An optimal replacement age of block replacement policy is determined which minimizes cost and nonoperating time when no information on prior distribution of parameters is given. Conclusion: We find the posterior distribution of parameters when lack of information on prior distribution, so that the optimal replacement age which minimizes the total cost and maximizes the total values is determined.

• 한국정보전자통신기술학회논문지
• /
• 제10권4호
• /
• pp.279-287
• /
• 2017

The aim of this study is to verify the difference in the level of happiness and preparation for old age according to the general characteristics of senior welfare center attendees. The subjects of the study were 148 attendees of OO senior welfare center in Seoul and a survey was conducted from July 4 to October 7, 2016. The method of analysis was by using the SPSSWIN 18.0 program, a 5% level of significance was verified in a frequency analysis, t-test, one-way ANOVA, correlation analysis, and multiple regression analysis. The results showed that first, from investigating the difference in the level of preparation for old age according to general characteristics, there was a significant difference according to age, occupation, monthly income, and physical condition but there was no significant difference according to gender and level of education. Second, as a result of observing the difference in the level of happiness according to general characteristics, although there was a significant difference according to occupation and monthly income, the difference level according to gender, age, level of education, health condition was less than a 5% level of significance. Third, from observing the correlation between the level of happiness and preparation of old age, it was found that happiness was positively correlated to preparation for old age in the following order: physical preparation for old age ((r=.576, p<.001), economic preparation for old age (r=.570, p<.001), psychological preparation for old age (r=.526, p<.001), social preparation for old age (r=.525, p<.001). It is forecasted that this study will be used as base line data for improvement of happiness and preparation for old age for the elderly according to general characteristics.

• 한국음향학회지
• /
• 제38권5호
• /
• pp.593-600
• /
• 2019

화자 간 음색의 유사성은 화자 인증 시스템의 성능을 하락 시킬 수 있는 요인이다. 본 논문은 화자 인증 시스템의 일반화 성능을 향상시키기 위해, 심층신경망에 멀티태스크 러닝 기법을 적용시켜 발화자의 화자 정보와 나이 정보를 함께 학습 시키는 기법을 제안한다. 멀티태스크 러닝 기법은 은닉층들이 하나의 태스크에 과적합 되지 않도록 하여 심층신경망의 일반화 성능을 향상시킨다고 알려져 있다. 하지만 심층신경망을 멀티태스크 러닝 기법으로 학습시키는 과정에서, 나이 정보에 대한 학습이 효율적으로 수행되지 않는 것을 실험적으로 확인하였다. 이와 같은 현상을 방지하기 위해, 본 논문에서는 심층신경망의 학습 과정 중 화자 식별과 나이 추정 목적 함수의 가중치를 동적으로 변경 하는 기법을 제안한다. 동일 오류율을 기준으로 RSR2015 평가 데이터세트에 대해 화자 인증 성능을 평가한 결과 나이 정보를 활용하지 않은 화자 인증 시스템의 경우 6.91 %, 나이 정보를 활용한 화자 인증 시스템의 경우 6.77 %, 나이 정보를 활용한 화자 인증 시스템에 가중치 변경 기법을 적용한 경우 4.73 %의 오류율을 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권4호
• /
• pp.1721-1737
• /
• 2020

Human apparent age estimation has become a popular research topic and attracted great attention in recent years due to its wide applications, such as personal security and law enforcement. To achieve the goal of age estimation, a large number of methods have been pro-posed, where the models derived through the cumulative attribute coding achieve promised performance by preserving the neighbor-similarity of ages. However, these methods afore-mentioned ignore the geometric structure of extracted facial features. Indeed, the geometric structure of data greatly affects the accuracy of prediction. To this end, we propose an age estimation algorithm through joint feature selection and manifold learning paradigms, so-called Feature-selected and Geometry-preserved Least Square Regression (FGLSR). Based on this, our proposed method, compared with the others, not only preserves the geometry structures within facial representations, but also selects the discriminative features. Moreover, a deep learning extension based FGLSR is proposed later, namely Feature selected and Geometry preserved Neural Network (FGNN). Finally, related experiments are conducted on Morph2 and FG-Net datasets for FGLSR and on Morph2 datasets for FGNN. Experimental results testify our method achieve the best performances.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2014년도 추계학술대회
• /
• pp.237-240
• /
• 2014

현재 우리가 살고 있는 정보화시대가 젊은이를 대상으로 급속히 발전하고 있으나 고령층에 대한 고려가 빈약한 실정이다. 2012년을 기준으로 고령층 인구(만55세~79세)가 천만명을 넘게 진입한 상태이다. 이러한 변화에 따른 문제 중 고령층 정보화교육에 따른 정보격차 및 일자리 창출이 사회적인 과제로 떠오르게 되었다. 이에 고령층에 대하여 전문적인 정보화교육을 시켜서 멘토로 하는 다양한 정보격차를 줄이는 학습방법을 제안 해 보기로 한다. 숙련된 멘토로부터 일자리 창출을 확립시키므로 사회의 일원으로서 보람과 사회경제적 도움으로 인간다운 삶을 영위하도록 함을 목표로 삶는다. 이에 양극화로 인한 사회적 갈등을 해소하고, 고령인구의 사회경제적 활동과 인간다운 삶을 위하여 정보격차 해소 및 고령층 일자리 창출은 반드시 필요하다. 따라서 본 연구는 한국정보화진흥원에서 실시하고 있는 어르신IT봉사단 활동을 기반으로 연구를 하였다. 대한민국의 초 고령화 시대를 맞이하는 과정에서 노인정보화 교육을 통한 정보격차 해소와 컴퓨터 재능기부를 활용한 노인 일자리 창출에 대한 방향제시가 연구의 목적이다.

• 한국도서관정보학회지
• /
• 제28권
• /
• pp.479-503
• /
• 1998

This paper examines the effects of the Internet on the librarianship and how the changing roles of libraries and librarians are reshaping the way they serve the users in the information age. Each area of operations and services in a so-called Digital/Electronic Library, which is a phenomenon occurring now not sometime later in the future, is reviewed. The importance of keeping in mind the library' main mission of providing a service, i.e. access to information, is stressed as the pivotal point of librarianship in this global society influenced by the presence of the Internet.

• 패션비즈니스
• /
• 제11권1호
• /
• pp.97-107
• /
• 2007

The purpose of this study was to examine the relationship structure in discount store focusing on age. Subjects for this study were 360 customers who had purchased fashion products in discount stores. For date analysis, $x^2$-test and regression analysis were used. As the result, when comparing groups by the age, women in their 20s, 30s and 40s had 'relationship maintenance intention' through satisfaction, trust and commitment. In the case of women in their 20s group, the information, reputation, product salesperson and price variables had the effects on satisfaction. The information, location, salesperson variable had direct effect on relationship maintenance. In the case of women in their 30s group, the benefits, information, reputation and price variables had the effects on satisfaction. And the information, salesperson variable had direct effect on relationship maintenance. In the case of women in their 40s group, the information, location and price variables had the effects on satisfaction. The reputation variable had direct effect on relationship maintenance. Based on these results, fashion marketing strategies of discount store would be suggested.

• 한국의류산업학회지
• /
• 제14권1호
• /
• pp.48-55
• /
• 2012

The purpose of this study was to review the results of previous researches related to cognitive age and to verify variables related to cognitive age. Using the key words of cognitive age, subjective age, self-perceived age and age difference from major academic databases in Korea (KISS), the related research articles were collected and analyzed. The results of this study were as follows: First, educational level and marital status were related to cognitive age but its relationship to the other demographic variables were inconsistent. Second, self-esteem and life satisfaction were associated with cognitive age. Third, consumer behavior such as information-seeking behavior, impulsive buying orientation and brand sensitivity and clothing behavior such as apparel shopping orientation and fashion leadership were related to cognitive age.

• 품질경영학회지
• /
• 제35권1호
• /
• pp.136-141
• /
• 2007

For many products failures depend on age and usage and, in this case, failures are random points in a two-dimensional plane with the two axes representing age and usage. Models play an important role in decision-making. In this research, an accelerate failure test (AFT) model is proposed to deal with the two-dimensional data. The parameters are proposed to be estimated through maximum likelihood estimators.

• 천문학회보
• /
• 제36권2호
• /
• pp.103.2-103.2
• /
• 2011

We measure the degree of polarization of the lunar regolith to map the distributions of the age and the particle size. We use a 12cm refracting telescope with a 2k-square pixel color CCD (R band) and a polarization filter. The angular resolution obtained is 3.02 km/pixel. Our goal is to obtain a map of the lunar particle size distribution on the lunar regolith and then that of the age distribution. Polarization of the light scattered by lunar surface contains information on their mean particle size. The mean particle size of the lunar surface has been decreased by continued micro-meteoroid impact over a long period. One can estimate the age of the lunar surface if the mean particle size is known. Particle sizes can be measured through observations of polarization because the mean particle size is related to the maximum polarization and albedo. The age and the particle size of the lunar regolith can give vital information for the future lunar exploration.