• Journal of KIISE:Software and Applications
• /
• v.26 no.11
• /
• pp.1261-1269
• /
• 1999

다층 퍼셉트론은 다양한 응용 분야에 성공적으로 적용되고 있는 대표적인 신경회로망 모델이다. 그러나 다층 퍼셉트론의 학습에서 나타나는 플라토에 기인한 느린 학습 속도와 지역 극소는 실제 응용문제에 적용함에 있어서 가장 큰 문제로 지적되어왔다. 이 문제를 해결하기 위해 여러 가지 다양한 학습알고리즘들이 개발되어 왔으나, 계산의 비효율성으로 인해 실제 문제에는 적용하기 힘든 예가 많은 등, 현재까지 만족할 만한 해결책은 제시되지 못하고 있다. 본 논문에서는 다층퍼셉트론의 베이시스 함수로 사용되는 시그모이드 함수를 보다 일반화된 형태로 정의하여 사용함으로써 학습에 있어서의 플라토를 완화하고, 지역극소에 빠지는 것을 줄이는 접근방법을 소개한다. 본 방법은 기존의 변형된 가중치 수정식을 사용한 학습 속도 향상의 방법들과는 다른 접근 방법을 택함으로써 기존의 방법들과 함께 사용하는 것이 가능하다는 특징을 갖고 있다. 제안하는 방법의 성능을 확인하기 위하여 간단한 패턴 인식 문제들에의 적용 실험 및 기존의 학습 속도 향상 방법을 함께 사용하여 시계열 예측 문제에 적용한 실험을 수행하였고, 그 결과로부터 제안안 방법의 효율성을 확인할 수 있었다. Abstract A multilayer perceptron is the most well-known neural network model which has been successfully applied to various fields of application. Its slow learning caused by plateau and local minima of gradient descent learning, however, have been pointed as the biggest problems in its practical use. To solve such a problem, a number of researches on learning algorithms have been conducted, but it can be said that none of satisfying solutions have been presented so far because the problems such as computational inefficiency have still been existed in these algorithms. In this paper, we propose a new learning approach to minimize the effect of plateau and reduce the possibility of getting trapped in local minima by generalizing the sigmoidal function which is used as the basis function of a multilayer perceptron. Adapting a new approach that differs from the conventional methods with revised updating equation, the proposed method can be used together with the existing methods to improve the learning performance. We conducted some experiments to test the proposed method on simple problems of pattern recognition and a problem of time series prediction, compared our results with the results of the existing methods, and confirmed that the proposed method is efficient enough to apply to the real problems.

• 한국사회정책
• /
• v.25 no.1
• /
• pp.125-159
• /
• 2018

The purpose of this study is to analyze the factors affecting service use intention of long-term care among the disabled. This study conducts a longitudinal study using the modified Andersen model by the disabled age groups of the 50-64 group and the 65 over group. This study examined random effects panel logit analysis for the 50-64 group and the 65 over group by age variations. The results have shown that there are different factors Influencing factors for each specific age group of the disabled(50-64, 65 over). The results were as follows: there are differences of factors between the 50-64 group and the 65 over group. This study found that predisposing factors of the service use intention of long-term care were significantly related to age, residential area, education status, existence of spouses. Enabling factors of the service use intention of long-term care were significantly related to long-term care service recognition, saving, personal salary income level, housing status. Need factors of the service use intention of long-term care were significantly related to chronic diseases, psychological health, IADL. So, there are a need for a policy considerations such as service for the mid-old age disabled by age groups. Finally, implications and future research directions were discussed based on the finding of the study.

• Journal of vocational education research
• /
• v.37 no.5
• /
• pp.55-78
• /
• 2018

The purpose of this study is to investigate the recognition of the professionalism required for strengthening the effective career guidance capacity of the secondary career teachers who are required to prepare for the coming fourth industrial revolution era. Based on these research objectives, we derived the required roles(8), the required competencies(20), and the contents(23) for enhancing professionalism by the required competencies, based on this, the questionnaire was composed and 217 respondents were collected and analyzed. First, the t-test was conducted to confirm the statistically significant difference between the current level and the important level of each content item by each role. As a result, it was found that in all roles except role of 'administrator' The t-value is statistically significant and the t-value distribution is high. Second, the demand value calculation and the priority ranking using the Borich demand calculation formula were found, and as a result, the directionality between the t value and the Borich demand was in agreement. Third, as a result of prioritizing using the Locus for Focus model, the contents of all 5 (middle school 7, high school 2) education contents were given priority. Fourth, three middle schools and five high schools were derived from the subordinate. Finally, we confirmed the relevance of the contents of education to actual educational necessity. Although this study has many limitations and limitations due to the fact that there are few prior data due to the segmentation of the subject related to the 4th Industrial Revolution and career guidance, it is necessary to develop educational training program I hope to be able to use it as basic data of various follow-up studies and make some suggestions.

• Journal of vocational education research
• /
• v.37 no.1
• /
• pp.1-29
• /
• 2018

The purpose of this study is to examine how the nursing teacher in vocational high school operates the class and examine various perceptions about the operation education. Moreover, This is to comparing the difference between nursing experience and education major about the operation the class. Based on the above findings, First, the nursing teacher should be recognized as a professional subject teacher and should have a qualification system that is appropriate as a professional subject teacher. Second, a formal education program should be established so that those who worked in the nursing field can become teachers of nursing education. It is recommended that nursing education within the graduate school of education be newly established and operated only for nursing graduates who are majored in the same country as developed countries. Third, if the establishment of qualifications and qualifications system can not be realized in a short time, it is necessary to systematically prepare the training for specialization nursing teachers only.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.13 no.6
• /
• pp.27-38
• /
• 2018

The purpose of this study is to find out common investment decision factors for CVC's invested technology-based startups and analyze them. We examined 17 CVCs that invested in technology startups for three years and six months from 2015 to June 2018. As a result, the final 9 CVCs that can be used in this study were confirmed and 188 companies were analyzed. This study was conducted as a case study to propose and demonstrate CVC investment objectives and investment decision factors analysis model. The results of this study are as follows. First, CVC focused on strengthening investment. Second, In 2015, Invested in an average of 19 months of technology-based startups. In recent years, we invested in 36 months of proven technology-based startups. Thirdly, ICT service was the main business type of the invested startups. Fourth, the investors were concentrated on the stage of Series A~B. It is observed that CVC investment determinants have a significant impact on product or service and parent company relations. In addition, it was found that factors such as innovation, business planning competency, enterprising, strategic competency, leadership, and opportunity recognition competency were influential factors for the startups of invested companies and it was found that these factors are important for CVC investment decision. Understanding of CVC investment determinants presented in this study is based on the establishment of the investment process of the investee, entrepreneurship and management education program. The results of this study can be applied to the selection of excellent startups, entrepreneurship education programs, mentoring, development of coaching guidelines, and establishment of investment process of other investment institutions when investing in CVC.

• Journal of Broadcast Engineering
• /
• v.23 no.6
• /
• pp.855-865
• /
• 2018

Sound event detection is one of the research areas to model human auditory cognitive characteristics by recognizing events in an environment with multiple acoustic events and determining the onset and offset time for each event. DCASE, a research group on acoustic scene classification and sound event detection, is proceeding challenges to encourage participation of researchers and to activate sound event detection research. However, the size of the dataset provided by the DCASE Challenge is relatively small compared to ImageNet, which is a representative dataset for visual object recognition, and there are not many open sources for the acoustic dataset. In this study, the sound events that can occur in indoor and outdoor are collected on a larger scale and annotated for dataset construction. Furthermore, to improve the performance of the sound event detection task, we developed a dual CNN structured sound event detection system by adding a supplementary neural network to a convolutional neural network to determine the presence of sound events. Finally, we conducted a comparative experiment with both baseline systems of the DCASE 2016 and 2017.

• Journal of the Korea Computer Graphics Society
• /
• v.25 no.3
• /
• pp.11-20
• /
• 2019

In this study, a novel virtual reality (VR) experience environment is proposed for enabling walking adaptation of visually impaired people. The core of proposed VR environment is based on immersive walking interactions and deep learning based braille blocks recognition. To provide a realistic walking experience from the perspective of visually impaired people, a tracker-based walking process is designed for determining the walking state by detecting marching in place, and a controller-based VR white cane is developed that serves as the walking assistance tool for visually impaired people. Additionally, a learning model is developed for conducting comprehensive decision-making by recognizing and responding to braille blocks situated on roads that are followed during the course of directions provided by the VR white cane. Based on the same, a VR application comprising an outdoor urban environment is designed for analyzing the VR walking environment experience. An experimental survey and performance analysis were also conducted for the participants. Obtained results corroborate that the proposed VR walking environment provides a presence of high-level walking experience from the perspective of visually impaired people. Furthermore, the results verify that the proposed learning algorithm and process can recognize braille blocks situated on sidewalks and roadways with high accuracy.

• Journal of the Korea Convergence Society
• /
• v.12 no.4
• /
• pp.237-248
• /
• 2021

Based on sports known to reduce adolescents' smartphone addiction, this study aims to identify the structural causal relationship between sport commitment and smartphone addiction. First of all, based on prior studies, it was intended to show that adolescents' sport commitment is effective in reducing smartphone addiction through social support and ego-resilience, and that such causal relationships vary significantly depending on gender. To this end, a survey of 605 middle and high school students was conducted, and the structural relationship between the four established latent variables was analyzed by analyzing the structural equation model using the SPSS 25.0 and AMOS 23.0 programs. According to the analysis, adolescents' sport commitment was found to have an adverse effect on smartphone addiction, and social support and ego-resilience in the relationship between the two variables were found to have a mediated effect. In addition, these results showed significant differences in gender. This study expanded the field of smartphone addiction and sports convergence study by revealing the relationship between sport commitment as psychological recognition variable and addiction Unlike previous studies using a participation time, duration, and intensity, and suggested that structural relationships with personal characteristic variables and gender-specific approaches are important. Through this, it is meaningful that it provided basic data for in-depth convergence study in the smartphone addiction and development of gender-specific programs.

• International Area Studies Review
• /
• v.14 no.2
• /
• pp.107-129
• /
• 2010

This study does the logistic regression analysis in a survey how Korean Small and Medium Enterprises(SMEs) recognize the effects of the Korea-India Comprehensive Economic Partnership (CEPA) on their business activities and strategies. The results show that Korean SMEs which have relatively more employees and bigger size in terms of sales have interest in the Korea-India CEPA. In particular, the SMEs which are engaged in their business activities with India have more interest in the CEPA than those that have not any business relations with India. However, the results show that they will not enter or expand their business fields in India even though the CEPA starts. It implies that Korean SMEs recognize that the CEPA will not affect their business significantly. Thus, this study has a policy implication for government to adopt/establish policies for SMEs utilizing the CEPA. It includes information services about the CEPA, India markets, a political corporation between the Governments for improving the partnerships in the trade and FDI and so on.

• Resources Recycling
• /
• v.30 no.3
• /
• pp.3-17
• /
• 2021

Currently, there is a drain of natural resources, environmental contamination, generation of waste, and problems of the earth's climate by CO₂ emissions according to mass production and overconsumption of mankind. It is effectuated by a linear economy that involves manufacturing products, use, and waste repeatedly; there is no guarantee in the lives of humans and the future of the globe if we do not find alternative proposals. For a sustainable developing society and to overcome the present global problems, we must successively change to a circular economy from a linear economy. The circular economy has the concept of an extended value chain in recovery, reuse, remanufacturing, and recycling, instead of discarding after the use of manufactured goods. New business models of circular economy have been realized to save the earth ecology and sustainable developing society in serious recognition of the linear economy system. New business models are established by creating a vision and developing a program, and by renovating technology, law, and financial support through a worldwide government policy.