• 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.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.634-634
• /
• 2015

최근 국지성 호우와 잦은 태풍으로 인한 돌발홍수가 빈번하게 발생하여 도심지에서의 호안유실과 범람으로 많은 외수침수의 피해가 발생하고 있다. 또한 기후변화에 따른 강우량의 증가와 집중호우로 인한 홍수 피해는 지속적으로 증가할 것으로 예상됨에 따라 대하천 유역을 중심으로 홍수범람예측 연구가 활발히 진행되고 있지만 대하천을 제외한 지방 중소하천의 연구가 미비한 실정이다. 이에 본 연구에서는 지방 중소하천 중 태풍과 집중호우의 영향이 많은 제주지역의 주요 하천 중의 하나인 한천 유역을 테스트베드로 선정하여 연구를 진행하였다. 한천은 강우 시에만 유출이 발생하는 건천으로, 집중호우 시 암반하상 조건, 복개, 교각 등으로 수위가 국부적으로 급격히 상승하는 경우가 있었다. 그리고 한천 하류부에는 도심이 위치하고 있어 돌발홍수 발생 시 막대한 피해가 발생한다. 이에 따라 홍수 피해를 줄이기 위한 제도화, 정책결정 등의 구조적 해결방안과 홍수 피해의 규모와 원인을 분석하는 비구조적 해결방안에 대한 연구가 시급하다. 따라서 본 연구에서는 홍수범람 등으로 인한 홍수 피해규모를 산정하여 각 정부부처 및 유관기관, 지자체에서 빠른 정책결정을 내릴 수 있는 자료를 제공하는 목적으로 제주도의 특성을 고려한 홍수범람위험도 산정 알고리즘을 개발하고자 한다. 본 연구에서는 제주 한천유역의 단면 자료와 빈도별 홍수량 자료를 이용하여 HEC-RAS 모형으로 수리학적 흐름특성 모의를 실사하였다. 모의된 결과를 바탕으로 ArcGIS 소프트웨어인 ESRI사의 ArcMap을 이용하여 빈도별 홍수위 자료와 제주지역 수치표고모형 자료를 활용한 빈도별 홍수범람지도를 산정하고, 좌안과 우안의 제방고로부터 위험도를 산정하여 홍수범람위험도를 각각 구축하였다. 구축된 결과를 이용하여 분석하고자하는 해당 빈도의 홍수위와 홍수량이 발생할 때의 피해지역을 예측하였으며, 예측된 지역과 제주시의 공시지가 자료를 중첩하여 피해지역에 대한 피해액을 산정하였다. 본 연구의 알고리즘을 적용한 2007년 태풍 '나리' 사상의 경우와 비교한 결과, '나리' 사상의 침수 흔적도와 유사한 홍수범람지도를 획득 할 수 있었으며, 모의된 유역의 하천 복개구간을 중심으로 홍수범람이 발생한다는 점과 우안보다 좌안에서의 홍수범람위험도와 피해액이 더 크게 나타난 점 등의 홍수범람 특성을 파악할 수 있었다. 본 연구에서 제시된 기법을 이용할 경우, 홍수에 의한 취약지에 대한 제방 설계 강화, 하천의 보수 정비 등 정책적 결정에 사용될 수 있을 것이며, 실시간 자료제공, 재해정보시스템 등에 적용하여 홍수범람 피해를 줄일 수 있는 기반기술이 될 것으로 사료된다.

• Geophysics and Geophysical Exploration
• /
• v.1 no.1
• /
• pp.57-63
• /
• 1998

The electrical resistivity method, one of old and widely used geophysical prospecting methods, has extended its scope to civil & environmental engineering areas. The electrical resistivity imaging technique was performed at the waste disposal site located in Junju to verify the applicability to the environmental engineering area. The dipole-dipole array, with the dipole spacing of 10 m, was applied along eight survey lines. The field data were obtained under the control of automatic acquisition softwares and topographic effects were corrected during processing stage. The processed resistivity images show that very low resistivity develops inside the disposal site and the distribution of low resistivity is exactly in accord with the boundary of the site except the river side. The depth of low resistivity zones is deeper toward the river side, which is interpreted that there is a high possibility for contaminants to be scattered to the river. From resistivity images, it was feasible to deduce the depth of waste disposal as well as the horizontal/vertical distribution of the contaminated zone, which proved the applicability of the electrical resistivity imaging technique to the environmental engineering area.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.11
• /
• pp.433-440
• /
• 2019

Recently, machine learning and data mining have been used for many disease prediction and diagnosis. Chronic diseases account for about 80% of the total mortality rate and are increasing gradually. In previous studies, the predictive model for chronic diseases use data such as blood glucose, blood pressure, and insulin levels. In this paper, world's first research, verifies the relationship between dyslipidemia and facial characteristics, and develops the predictive model using machine learning based facial characteristics. Clinical data were obtained from 5390 adult Korean men, and using hypertriglyceridemia and facial characteristics data. Hypertriglyceridemia is a measure of dyslipidemia. The result of this study, find the facial characteristics that highly correlated with hypertriglyceridemia. FD_43_143_aD (p<0.0001, Area Under the receiver operating characteristics Curve(AUC)=0.652) is the best indicator of this study. FD_43_143_aD means distance between mandibular. The model based on this result obtained AUC value of 0.662. These results will provide a basis for predicting various diseases with only facial characteristics in the screening stage of disease epidemiology and public health in the future.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.5
• /
• pp.919-939
• /
• 2021

As a next-generation communication technology, 5G networks are capable of handling large amounts of traffic based on higher speeds, shorter communication delays, and higher connectivity compared to 4G networks. In this 5G network environment, base stations are installed all over the city at high density due to their characteristics, and are connected to user terminals to provide services. Therefore, if the base station is damaged by a malicious attacker, it is expected to cause great damage to users and society as a whole. So the need for secure communication equipment such as 5G base stations has emerged. Therefore, in this paper, we propose the security functional requirements derived using threat modeling, a systematic methodology for 5G base stations, and the security assurance requirements at the level that can cope with the backdoor issues. The security requirements proposed in this paper can be used for base station design and development to construct a secure network environment as a security evaluation standard for 5G base stations.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.49 no.2
• /
• pp.1-16
• /
• 2021

The act of design in landscape architecture consists of a concept within a designer's mind, technical representations, and finally, a process of construction. In the 4th Industrial Revolution, the design process is facing many changes due to the rapid development of computer technology and the IT ecosystem. Computer technology was initially developed for simple functions, such as mathematical calculation and graphic representation. However, after the spread of Personal Computers, starting with IBM and Macintosh, programming languages and hardware rapidly developed, algorithms and applications became specialized, and the purpose of using computers became very diverse. This study diagnoses issues concerning the functions and roles that new design methods, such as computational design, parametric design, and algorithmic design, can play in landscape architecture based on changes in the digital society. The study focused on the design methodology using parametric technology, which has recently received the most attention. First, the basis for discussion was developed by examining the main concepts and characteristics of parametric design in modern landscape architecture. Prior research on the use of parametric design in landscape architecture was analyzed, as were the case studies conducted by landscape design firms. As a result, it was confirmed that parametric design has not been sufficiently discussed in terms of the number and diversity of studies compared to other techniques investigated by landscape design firms. Finally, based on the discussion, the study examined specific cases and future possibilities of the parametric design in landscape architecture.

• Land and Housing Review
• /
• v.12 no.2
• /
• pp.61-78
• /
• 2021

This study aims to understand the thematic trends globally developed in the 'Green Urbanism' related research. Research methodology is based on systemic review of international literature published for the past 20 years period between 2000 and 2020. The specific methods applied include not only literature search by citation, co-authorship, and co-occurrence but social network analysis in order to find correlations among the publication. The correlations are visualized and analysed using VOSviewer and Ucinet software. The analysis indicates that total of 51 studies were carried out by 89 authors from 54 institutions across 21 countries during the period. The majority of the research was done by a country-specific study and only a few research were collaborative studies with other countries. The most common theme that occurred in the early years was 'sustainability and the theme evolved toward specific ones such as 'built environment', 'infrastructure', and 'health'. Having considered that climate change has become a global challenge, green urbanism is expected to be a future direction to pursue environmentally sustainable urban spaces. This study also implies that governance, policy support, and intervention are crucial factors in developing sustainable urban spaces.

• The Journal of the Korea Contents Association
• /
• v.21 no.7
• /
• pp.363-373
• /
• 2021

The purpose of this research is to analyze the Spill-over economic effect of the cultural and creative industries(CCI) in Henan Province, China. The research object is the CCI of Henan Province, which is mainly based on five sectors out of 42 industries in the industrial association table of the Statistical Bureau of Henan Province, China in 2017 (culture, sports; recreation and research sector; experimental development and integrated technical services sector; information transmission, computer services and software sector; education sector, etc), and is analyzed through secondary integration and redefinition of the CCI of Henan Province. Through the analysis of Henan Province Industry Association Table, this paper provides some enlightenment to the future direction of the cultural and creative industries. The main analysis results are as follows. The total production inducement of the CCI in Henan province is 48,848 billion yuan, and in particular, the production inducement coefficient of the industry in Henan province is 2.72809, 2.23909 (total of columns and rows), Index of the power of dispersion is 0.26325, and the index of the sensitivity of dispersion is 0.87535. Income induction coefficient is 0.55211, production tax induction coefficient is 0.09291. Because CCI of Henan Province has full development potential, the government needs to provide active support and policy support, in addition to the need for legal provisions and supervision of market management. In order to improve the innovative development of the CCI, it is necessary to develop a new model of "CCI+X".

• Tunnel and Underground Space
• /
• v.29 no.3
• /
• pp.172-183
• /
• 2019

The generalized Hoek-Brown (GHB) failure criterion developed by Hoek et al. (2002) is a nonlinear function which defines a stress condition at failure of rock mass. The relevant strength parameter values are systematically determined using the GSI value. Since GSI index is a value quantifying the condition of in-situ rock mass, the GHB criterion is a practical failure condition which can take into the consideration of in-situ rock mass quality. Considering that most rock mechanics engineers are familiar with the linear Mohr-Coulomb criterion and that many rock engineering softwares incorporate Mohr-Coulomb criterion, the equations for the equivalent friction angle and cohesion were also proposed along with the release of the GHB criterion. The proposed equations, however, fix the lower limit of the minor principal stress range, where the linear best-fitting is performed, with the tensile strength of the rock mass. Therefore, if the tensile stress is not expected in the domain of analysis, the calculated equivalent friction angle and cohesion based on the equations in Hoek et al. (2002) could be less accurate. In order to overcome this disadvantage of the existing equations for equivalent friction angle and cohesion, this study proposes the analytical formula which can calculate optimal equivalent friction angle and cohesion in any minor principal stress interval, and verified the accuracy of the derived formula.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.5
• /
• pp.512-520
• /
• 2019

It is quite useful to represent speeches visually for learners who study foreign languages as well as the hearing impaired who cannot directly hear speeches, and a number of researches have been presented in the literature. They remain, however, at the level of representing the characteristics of speeches using colors or showing the changing shape of lips and mouth using the animation-based representation. As a result of such approaches, those methods cannot tell the users how far their pronunciations are away from the standard ones, and moreover they make it technically difficult to develop such a system in which users can correct their pronunciation in an interactive manner. In order to address these kind of drawbacks, this paper proposes a speech visualization model based on the relative distance between the user's speech and the standard one, furthermore suggests actual implementation directions by applying the proposed model to the visualization of Korean vowels. The method extract three formants F1, F2, and F3 from speech signals and feed them into the Kohonen's SOM to map the results into 2-D screen and represent each speech as a pint on the screen. We have presented a real system implemented using the open source formant analysis software on the speech of a Korean instructor and several foreign students studying Korean language, in which the user interface was built using the Javascript for the screen display.