• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.4
• /
• pp.487-492
• /
• 2011

In this paper, we proposed a new design methodology to improve the classification performance of the Nearest Prototype Classifier which is one of the simplest classification algorithm. To optimize the position vectors of the prototypes in the nearest prototype classifier, we use the differential evolutionary algorithm. The optimized position vectors of the prototypes result in the improvement of the classification performance. The new method to determine the class labels of the prototypes, which are defined by the differential evolutionary algorithm, is proposed. In addition, the experimental application covers a comparative analysis including several previously commonly encountered methods.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.3
• /
• pp.67-75
• /
• 2016

This paper investigates how artificial individuals with behavioral memory circuits adaptively evolve with respect to given environments on a cell-level simulation framework simulating artificial individuals. This makes it possible for us to analyse the advantages of artificial individuals with behavioral memory circuits against the simple artificial individuals that can do only simple reactions with respect to the environments and to know which advanced reactions are possible. In order to do this analysis, we experimented various tests on a specific prey pattern and examined the results. As a first experiment, we tested that artificial individuals with four memory steps competed against from those without memory step to those with three memory steps. Experimental results showed that the artificial individuals with four memory steps were superior to most others. However, artificial individuals with two memory steps were better than those with four memory steps. This was caused that the artificial individuals with two memory steps could evolve faster than those of four memory steps. In a second experiment that all types of artificial individuals are simultaneously evolved, the artificial individuals with two memory steps also showed the best result in the experiment. We could conclude that the artificial individuals with memory was better than those without memory and the best memory steps of artificial individuals were depended on the complexity of prey patterns.

• Proceedings of the KAIS Fall Conference
• /
• 2012.05a
• /
• pp.127-129
• /
• 2012

대중음악산업에서 이용매체의 변화는 음악 소비자들의 이용패턴의 변화로 나타나며, 여기에 TV 등 매크컴의 복합적인 영향이 음악 소비자들의 선호도에 추가적인 영향을 미친다. 본 논문에서는 소유에서 소비로 이미 빠르게 진화하고 있는 국내 음악 소비자들의 이용패턴을 살펴보고 앞으로의 추세를 살펴 보고자 하였다.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.20 no.1
• /
• pp.211-224
• /
• 2018

It is important to design tunnel support system considering the various loads acting on the tunnel because they have a direct impact on the stability of tunnels. In Korea, standardized support patterns are defined based on the rock mass classification system depending on the project, and it is stated that it should be modified appropriately considering the behavior of tunnel during construction. In this study, the tunnel support pattern optimization method is suggested based on the convergence-confinement method, earth pressure, axial force of rock bolt, and moment acting on the shotcrete. The length and spacing of the rock bolts and the thickness of the shotcrete were optimized by using the differential evolution algorithm (DEA) and the results were compared to the standard support pattern III for railway tunnel. Rock bolt length can be reduced and the installation interval can be widened for shallow tunnel. As the depth of tunnel increases, the thickness of shotcrete increases linearly. Therefore, the thickness of shotcrete should be thicker than the standard support pattern as the depth of tunnel increases to secure the stability of tunnel.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1991.06a
• /
• pp.3-6
• /
• 1991

본 연구에서는 음성의 패턴작성법을 개선하고 음성인식율을 향상시키기 위하여 퍼지패턴매칭을 개선한 뉴럴퍼지패턴매칭에(a neural-fuzzy pattern matching)의해 특정화자 고립단어인식을 수행하였다. 이 방법에서는 신경회로망의 연상기억에 의한 사상에 의해 패턴을 작성하여 주파수변동을 흡수하고 표준패턴고 선형매칭에 의해 유사도를 측정하여 인식하므로써 시간변동의 문제를 보완하였다. 또한, 이 방법에서 사용하는 특징피라미터는 2진화 스펙트럽이며, 유사도는 논리연산에 의해 측정되기 때문에 종래의 왜곡척도를 이용한 DTW 방법에 비해 기억용량과 계산량이 매우 작다. 이 방법의 인식성능을 평가하기 위하여 남녀가 발성한 28개의 도시명을 대상으로 인식실험을 수행한 결과, 신경회로망을 이용하지 않은 퍼지패턴매칭보다 오인식을 감소시켰으며, 뉴럴-퍼지 패턴매칭에 의한 특정화자 고립단어인식의 우수성을 확인하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06a
• /
• pp.22-24
• /
• 2006

마이크로어레이는 전체 유전체 수준의 mRNA 발현 여부에 대한 측정이 가능하다는 점에서 분자생물학의 실험 도구로서 가장 강력한 도구 중에 하나로 부각되어 있다. 현재까지 마이크로어래이의 결과로부터 유사한 발현 패턴을 찾기 위한 여러 가지 바이클러스터링 알고리즘들이 개발되어 왔다. 하지만 대다수의 알고리즘들이 최적의 바이클러스터들을 찾기보다는 일정 수준의 가능한 바이클러스터의 결과만을 제시하고 있다. 본 논문에서는 다른 개체집단들과 상호 진화하는 공진화적 학습에 의한 진화연산 기법을 통하여 유전자-조건의 매트릭스로부터 열과 행을 동시에 클러스터링하는 공진화적 바이클러스터링 알고리즘(co-evolutionary biclustering algorithm: CBA)을 제안하고자 한다. CBA는 유전자발현 데이터에서 유전자-조건의 상호의존적인 부성분들로 구성된 최적화 문제에 적합한 계산방식이라고 할 수 있다. 인간 유전자 발현 데이터에 대한 실험 결과. 제시한 알고리즘은 이전의 알고리즘에 비해 발견한 바이클러스터의 패턴 유사도에 있어서 우수한 성능을 보이고 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.490-492
• /
• 2002

본 논문은 물 속 유충인 깔따구의 움직임을 관찰한 데이터에 Active Contour Model을 적용하여 깔따구 상태의 특징을 추출하는 방법을 제안한다. 1987년 소개된 Active Contour Model은 주어진 영상에 놓인 커브를 그 커브에 의해 분할된 영상의 에너지 값을 최소화하는 방향으로 진화하게 함으로써 영상 내 객체의 경계를 찾게 하는 영상분할 방법이다. Chan과 Vese에 의해 개선된 Model을 이용하여 다이아지논이 처리되기 전과 후의 깔따구 행동 패턴의 특징을 찾아낸다. 우선 깔따구의 움직임 궤적을 0.25초를 간격으로 관찰하여 구해진 속도벡터의 위상영상을 만든다.그리고 위상영상에 Active Contour를 두어 진화시키면서 시간에 따라 감소하는 에너지 값의 그래프에서 구해진 기울기로 깔따구 행동 패턴의 특징을 추출한다.

• Proceedings of the CALSEC Conference
• /
• 2005.11a
• /
• pp.138-146
• /
• 2005

• Journal of Science and Technology Studies
• /
• v.6 no.2 s.12
• /
• pp.1-44
• /
• 2006

This paper aims to analyse post catch-up innovation activities of Korean firms. Traditional approach focused on incremental innovation through fast catch-up process. Recently, Korean firms present new patterns of innovation activities. It includes the development of state-of-the-art frontier products in capital intensive manufacturing sector, the achievement of architectural innovation through adopting newly developed components and the creation of new market based on the basic technological capabilities. This paper attempts to formulate conceptual framework for analysing post catch-up innovation activities and to analyse stylized facts of post catch-up activities based on the case studies on technology based small and medium Korean firms.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.12
• /
• pp.10-19
• /
• 1999

Cellular Automata Neural Systems(CANS) are neural networks based on biological development and evolution. Each neuron of CANS has local connection and acts as a form of pulse according to the dynamics of the chaotic neuron. CANS are generated from initial cells according to the CA rule. In the previous study, to obtain the useful ability of CANS, we make the pattern of initial cells evolve. However, it is impossible to represent all solution space, so we propose an evolving method of CA rule to overcome this defect in this paper. DNA coding has the redundancy and overlapping of gene and is apt for the representation of the rule. In this paper, we show the general expression of CA rule and propose translation method from DNA code to CA rule. The effectiveness of the proposed scheme was verified by applying it to the navigation problem of autonomous mobile robot.