• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.249-251
• /
• 2021

PBFT (Practical Byzantine Fault Tolerant) is a consensus algorithm that guarantees higher processing speed compared to PoW (Proof of Work) and absolute finality that records are not overturned due to the superiority of computing power. However, due to the complexity of the message, there is a limit that the network load increases exponentially as the number of participating nodes increases. PBFT is an important factor in determining the performance of a blockchain network, but studies on evaluation metrics and evaluation technologies are lacking. In this paper, we propose a PBFT evaluation framework that is convenient to change the consensus algorithm to easily evaluate quantitative indicators and improved methods for evaluating PBFT.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.199-202
• /
• 2000

본 논문에서는 퍼지-유전자 접근방법을 이용한 무감독 특징 선택방법에 대하여 나타내었다. 이 방법은 각각의 특징들의 중요도에 따라 순서를 정하기 위해 사용되는 weighted distance 를 포함하는 특징 평가 지표 (feature evaluation index)를 최소화시키는데 있다. 또한 특징 평가 지표에서 사용되는 각 패턴들의 쌍에 대하여 근접함의 정도를 퍼지 멤버쉽 함수를 이용하여 결정하고 유전자 알고리즘은 평가 지표를 최소화시킴으로써 각 특징의 중요도를 나타내는 최적의 weighting 계수의 집합을 한기 위하여 적용하였다.

• Journal of KIISE:Software and Applications
• /
• v.31 no.1
• /
• pp.82-88
• /
• 2004

Comparing many programs like programming assignments one by one requires many costs. Moreover, if the checker would evaluate or grade assignments, much more time will be required. Even through the checker invest much time, fairness is not always guaranteed. These problems can be solved easily by grouping similar programs. So, programs after grouping can be easily evaluated and graded. In this paper, we propose and implement algerian performing grouping by similarity on many programs. The grouping algorithm evaluates similarity using algorithm proposed in (9), and performs a grouping following high similarity order. By using this grouping algorithm, the number of comparison among N programs can be reduced from N-1 times to N(N-1)/2 times. In the part of experiment and evaluation of this paper, we actually showed evaluation result by similarity using randomly 10 programming assignments at the university.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.171-176
• /
• 2020

Reinforcement learning is a well-known method for training an artificial software agent for a video game. Even though many reinforcement learning algorithms have been proposed, their performance was varies depending on an application area. This paper compares the performance of the algorithms when we train our reinforcement learning agent for a 2D racing game. We defined performance metrics to analyze the results and plotted them into various graphs. As a result, we found ACER (Actor Critic with Experience Replay) achieved the best rewards than other algorithms. There was 157% gap between ACER and the worst algorithm.

• Journal of KIISE:Databases
• /
• v.27 no.2
• /
• pp.199-208
• /
• 2000

The k-nearest neighbor search query based on similarity is very important for content-based multimedia information retrieval(MIR). The conventional exact k-nearest neighbor search algorithm is not efficient for the MIR application because multimedia data should be represented as high dimensional feature vectors. Thus, an approximate k-nearest neighbor search algorithm is required for the MIR applications because the performance increase may outweigh the drawback of receiving approximate results. For this, we propose a new approximate k-nearest neighbor search algorithm for high dimensional data. In addition, the comparison of the conventional algorithm with our approximate k-nearest neighbor search algorithm is performed in terms of retrieval performance. Results show that our algorithm is more efficient than the conventional ones.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.5
• /
• pp.1084-1090
• /
• 2015

In this paper, we propose a Tabu search algorithm to efficiently deploy the sensor nodes for maximizing the network sensing coverage in wireless sensor networks. As the number of the sensor nodes in wireless sensor networks increases, the amount of calculation for searching the solution would be too much increased. To obtain the best solution within a reasonable execution time in a high-density network, we propose a Tabu search algorithm to maximize the network sensing coverage. In order to search effectively, we propose some efficient neighborhood generating operations of the Tabu search algorithm. We evaluate those performances through some experiments in terms of the maximum network sensing coverage and the execution time of the proposed algorithm. The comparison results show that the proposed algorithm outperforms other existing algorithms.

• The KIPS Transactions:PartD
• /
• v.18D no.2
• /
• pp.101-110
• /
• 2011

According to the increasing use of blogs, a huge number of posts have appeared in a blogosphere. This causes web surfers to face difficulty in finding the quality posts in their search results. As a result, post ranking algorithms are required to help web serfers to effectively search for quality posts. Although there have been various algorithms proposed for web-page ranking, they are not directly applicable to post ranking since posts have their unique features different from those of web pages. In this paper, we propose post ranking algorithms that exploit actions performed by bloggers. We also evaluate the effectiveness of post ranking algorithms by performing extensive experiments using real-world blog data.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1374-1375
• /
• 2008

본 논문은 3상 일괄형 GIS의 부분방전 진단을 위해 새롭게 개발한 진단 알고리즘에 관한 것이다. 진단 알고리즘 개발을 위해, 먼저 실시간 부분방전 데이터를 행벡터 및 열벡터로 구성하고 각각의 벡터에서 통계 특징량 및 질감 특징량을 추출하였다. 다음으로 이들 특징량을 GA-NN(Genetic Algorithm - Neural Network) 학습에 적용하여 진단 알고리즘을 구성하였다. 또한 진단 알고리즘의 위상독립성은 부분방전 신호의 위상변화에 관계없이 진단결과가 일치하는 것을 확인함으로써 검증하였다. 개발한 진단알고리즘의 실증 평가를 위해, 부분방전이 발생되고 있는 국내 3상 일괄형 GIS 변전소에 적용하였다. 적용 결과, 위상에 관계없이 부분방전 발생원을 정확히 진단함을 확인하였고, 이를 통해 개발 알고리즘의 우수성을 입증하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.7
• /
• pp.11-17
• /
• 2010

This paper proposes an improved SMO solving a quadratic optmization problem for class imbalanced learning. The SMO algorithm is aproporiate for solving the optimization problem of a support vector machine that assigns the different regularization values to the two classes, and the prosoposed SMO learning algorithm iterates the learning steps to find the current optimal solutions of only two Lagrange variables selected per class. The proposed algorithm is tested with the UCI benchmarking problems and compared to the experimental results of the SMO algorithm with the g-mean measure that considers class imbalanced distribution for gerneralization performance. In comparison to the SMO algorithm, the proposed algorithm is effective to improve the prediction rate of the minority class data and could shorthen the training time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.10A
• /
• pp.1492-1498
• /
• 2000

Pilot symbol aided modulation (PSAM) using the conventional sinc interpolation (CSI) achieves nearly the same BER performance as Caver' optimal Wiener interpolation but with much less complexity. The CSI, however, has to use a non-rectangular window function that is applied to the sinc function to smooth out the abrupt truncation of rectangular window. In this paper, we propose the modified sinc interpolation (MSI). With the weighting factor the MSI scheme with no window has almost the same BER performance as the CSI scheme using window, In addition, if we use the MSI with a window its BER performance gets close to that of the theoretical one. We assume the multicarrier QAM system and an optimal frame structure for performance evaluation.