• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2002년도 가을 학술발표논문집 Vol.29 No.2 (1)
• /
• pp.223-225
• /
• 2002

Decision tree는 목표 데이터에 대한 계층적 관점을 보여준다는 의미에서 데이터를 보다 잘 이해하는데 많은 도움이 되나 탐욕법(greedy algorithm)에 의한 트리 생성법의 한계로 인해 최적의 예측자라고는 할 수가 없다. 이와 같은 약점을 보완하기 위하여 일반적 방법으로 생성한 decision tree에 대하여 다차원 연관규칙 알고리즘을 적용함으로써 짱은 길이의 최적 부분 규칙집합을 구하는 방법을 제시하였고 실험을 통해 그와 같은 사실을 확인하였다.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제25권4호
• /
• pp.149-161
• /
• 2021

This study proposes a methodology for mission/path planning of an unmanned aerial vehicle (UAV) using an artificial potential field with the Markov Decision Process (MDP). The planning problem is formulated as an MDP. A low-resolution solution of the MDP is obtained and used to define an artificial potential field, which provides a continuous UAV mission plan. A numerical case study is conducted to demonstrate the validity of the proposed technique.

• 정보학연구
• /
• 제8권1호
• /
• pp.51-56
• /
• 2005

The Back propagation algorithm is a very effective supervised training method for multi-layer feed forward neural networks. This paper studies the decision boundary formation based on the Back propagation algorithm. The discriminating powers of several neural network topology are also investigated against five manually created data sets. It is found that neural networks with multiple hidden layer perform better than single hidden layer.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2006년도 춘계학술대회 학술발표 논문집 제16권 제1호
• /
• pp.279-282
• /
• 2006

네트워크를 기반으로 한 컴퓨터 시스템이 현대 사회에 있어서 더욱 더 불가결한 역할을 하는 것에 따라, 네트워크 기반 컴퓨터 시스템은 침입자의 침입 목표가 되고 있다. 이를 보호하기 위한 침입탐지시스템(Intrusion Detection System : IDS)은 점차 중요한 기술이 되었다. 침입탐지시스템에서 패턴들을 분석한 후 정상/비정상을 판단 및 예측하기 위해서는 초기단계인 특징추출이나 선택이 매우 중요한 부분이 되고 있다. 본 논문에서는 IDS에서 중요한 부분인 feature selection을 Data Mining 기법인 Genetic Algorithm(GA)과 Decision Tree(DT)를 적용해서 구현했다.

• 한국정보통신학회논문지
• /
• 제1권2호
• /
• pp.199-207
• /
• 1997

The Maximum Likelihood Sequence Estimation scheme is modified to improve the error performance of the correlative coding in the Tamed FM. To remove intersymbol interference, the Decision Feedback Equalization scheme with the delayed LMS algorithm and the Viterbi algorithm(10-symbol delay) in the delayed adaptive equalization are proposed for the performance of decision-directed adaptive equalization under the High Frequency channels, and the condition of convergence is analyzed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권10호
• /
• pp.4007-4024
• /
• 2020

It is tremendously important to construct decision trees to use as a tool for knowledge representation from a given decision table. However, the usual algorithms may split the decision table based on each value, which is not efficient for numerical attributes. The methodology of this paper is to split the given decision table into binary groups as like the CART algorithm, that uses binary split to work for both categorical and numerical attributes. The difference is that it uses split for each attribute established by the directed acyclic graph in a dynamic programming fashion whereas, the CART uses binary split among all considered attributes in a greedy fashion. The aim of this paper is to study the effect of binary splits in comparison with each value splits when building the decision trees. Such effect can be studied by comparing the number of nodes, local and global misclassification rate among the constructed decision trees based on three proposed algorithms.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1346-1349
• /
• 2004

In this paper, we have proposed the new link-weight calculating function using for routing decision in WDM networks. The proposed link-weight calculating functions includes following factors those are available wavelengths per link, distance loss, total wavelengths, and limited wavelength conversion. The calculated link-weight is applied into the algorithm of routing decision in order to determine the available lightpath that qualifies user requests. The objective is to improve the performance of wavelengths assignment with fast determining the suitable lightpath by using the proposed link-weights calculating function. The analytical model of WDM switching networks is introduced for numerical analysis. The link-weight calculating function is performed. Finally, the performance of proposed algorithm is displayed with numerical results in term of the blocking probability, the probability that connection requests from users are rejected due to there are no available lightpath to be assigned for them. It is also shown that the blocking probability is varied in depending on the number of available wavelengths and the degree of wavelength conversion. The numerical results also show that the proposed link-weight calculating function is more cost-effective choice for the routing decision in WDM switching networks.

• 한국산학기술학회논문지
• /
• 제12권5호
• /
• pp.2390-2395
• /
• 2011

이 논문에서는, 통신 채널의 다중경로에 의한 심볼간 간섭 (ISI)과 와 임펄스 잡음을 극복하도록 하기 위해 오류 확률의 유클리드 거리를 최소화하는 결정 궤환 등화 알고리듬을 제안하였다. 오류 확률의 유클리드 거리는 오류 확률과 델타 함수의 차이로 정의하였다. 등화기 가중치에 대해 유클리드 거리를 최소화함으로써 제안한 알고리듬은 임펄스 잡음에 대한 강건성 뿐 아니라 심각한 다중경로 채널의 잔여 ISI를 제거하는 효과를 보였다.

• 한국정보전자통신기술학회논문지
• /
• 제3권3호
• /
• pp.73-80
• /
• 2010

상수 모듈러스 오차와 두확률 밀도 함수의 유클리드 거리에 기본을 둔 블라인드 알고리듬은 정보 이론적 학습 방법의 장점에도 불구하고 복소 채널의 위상 회전을 극복하지 못해 열악한 성능을 보인다. 이 논문에서는, 출력 전력이 다중 모듈러스 값의 근방에 있을 때 decision-directed 모드로 동작하는 기법을 정보 이론적 학습에 추가하므로서 복소 채널의 위상 회전 문제를 해결할 수 있음를 보였다. 복소 채널 모델과 16 QAM 방식에 대한 시뮬레이션 결과에서 복소 채널의 위상 회전 문제가 해결되어 현격한 성능 향상을 보였다.

• 경영과학
• /
• 제25권1호
• /
• pp.1-9
• /
• 2008

This paper investigates the R&D project portfolio selection problem. Despite its importance and impact on real world projects, there exist few practical techniques that help construct an non-dominated portfolio for a decision makers satisfaction. One of the difficulties constructing the portfolio is that such project portfolio problem is, in nature, a multi-attribute decision-making problem, which is an NP-hard class problem. This paper investigates the R&D project portfolio selection problem. Despite its importance and impact on real world projects, there exist few practical techniques that help construct an non-dominated portfolio for a decision makers satisfaction. One of the difficulties constructing the portfolio is that such project portfolio problem is, in nature, a multi-attribute decision-making problem, which is an NP-hard class problem. In order to obtain the non-dominated portfolio that a decision maker or a user is satisfied with, we devise a user-interface algorithm, in that the user provides the maximum/minimum input values for each project attribute. Then the system searches the non-dominated portfolio that satisfies all the given constraints if such a portfolio exists. The process that the user adjusts the maximum/minimum values on the basis of the portfolio found continues repeatedly until the user is optimally satisfied with. We illustrate the algorithm proposed, and the computational results show the efficacy of our procedure.