• Journal of Broadcast Engineering
• /
• v.13 no.1
• /
• pp.86-91
• /
• 2008

In the field of network data analysis, the research of how much the estimation data reflects the population data is inevitable. This paper compares and analyzes the well known MIT Lincoln Lab network data, which is composed of collectable standard information from the network with the KDD CUP 99 dataset which was composed from the MIT/LL data. For comparison and analysis, the protocol information of both the data was used. Correspondence analysis was used for analysis, SVD was used for 2 dimensional visualization and weigthed euclidean distance was used for network data quantification.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.29-30
• /
• 2019

본 논문에서는 다중 컨트롤러가 존재하는 분산 SDN 환경에서 과도한 제어 메시지로 인한 과부하된 컨트롤러의 부하를 줄이기 위하여 이주할 스위치를 K-means 군집화와 Harmony Search(HS)를 기반으로 선정 하는 기법을 제안하였다. 기존에 HS를 이용하여 이주할 스위치를 선택하는 기법이 제시되었으나, 시간 소모에 비하여 정확도가 부족한 단점이 있다. 또한 Harmony Memory(HM) 구축을 위해 메모리 소모 또한 크다. 이를 해결하기 위하여 본 논문에서는 유클리드 거리를 기반으로 하는 K-means 군집화를 이용하여 이주할 스위치를 골라내어 HM의 크기를 줄이고 이주 효율을 향상 시킨다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.26-26
• /
• 2022

본 연구는 하천중심선을 활용하여 Sentinel-1 위성영상기반 중소규모 하천 수체(水體) 추출 방법을 제안한다. 한강 유역의 한탄강 일부를 연구지역으로 선정하였으며, 이 지역을 촬영한 Sentinel-1 위성영상자료를 수집하였다. 여기에 개발한 방법의 검증을 위하여 유사시간대의 고해상도 광학위성 PlanetScope을 함께 수집하였다. 본 연구에서는 하천의 수체를 효과적으로 추출하기 위하여 국토지리정보원에서 제공하는 하천중심선 자료를 활용하였다. 하천중심선을 따라 유클리드 거리를 가중치로 산정한 자료(DST)와 Sentinel-1의 VH, VV 편광을 조합한 k-means 방법을 통해 위성영상의 픽셀을 군집화하였고, 최적의 매개변수 값을 산출하였다. 이 매개변수를 활용하여 Sentinel-1의 VV편광, VH편광 그리고 DST의 상관관계에 따른 타원방정식 형태의 계산식을 도출할 수 있었다. 수집한 자료의 검증결과 평균적으로 정확도는 0.65~0.75, kappa 계수는 0.8 내외를 보여 상당히 일치함을 확인할 수 있었다. 또한, 추가 확보한 30여 개의 Sentinel-1 위성영상을 제안 방법으로 추출한 수체의 면적과 유량 값을 비교해 본 결과, 유사한 변화 양상을 보였다. 본 연구는 하천 중심선자료를 활용하여 참값이 없더라도 수체 면적 추정이 가능함을 확인하였다. 제안한 방법은 현존하는 수체추출 방법보다 간단하고 신속하게 수체를 추출할 수 있을 것으로 보인다. 추후, 딥러닝을 통한 수체 식별을 추가 진행을 통해. 정확도를 높일 수 있을 것으로 기대한다.

• Annual Conference of KIPS
• /
• 2008.11a
• /
• pp.714-717
• /
• 2008

협업여과 추천기법에는 사용자 기반 협업여과와 아이템 기반 협업여과가 있으며, 절차는 유사도 측정, 이웃 선정, 예측값 생성 단계로 이루어진다. 유사도 측정 단계에는 유클리드 거리(Euclidean Distance), 코사인 유사도(Cosine Similarity), 피어슨 상관계수(Pearson Correlation Coefficient) 방법 등이 있고, 이웃 선정 단계에는 상관 한계치(Correlation-Threshold), 근접 N 이웃(Best-N-Neighbors) 방법 등이 있다. 마지막으로 예측값 생성 단계에는 단순평균(Simple Average), 가중합(Weighted Sum), 조정 가중합(Adjusted Weighted Sum) 등이 있다. 이처럼 협업여과 추천기법에는 다양한 기법들이 사용되고 있다. 따라서 본 논문에서는 사용자 기반 협업여과와 아이템 기반 협업여과 추천기법에 사용되는 유사도 측정 기법과 예측값 생성 기법의 최적화된 조합을 알아보기 위해 성능 실험 및 비교 분석을 하였다. 실험은 GroupLens의 MovieLens 데이터 셋을 활용하였고 MAE(Mean Absolute Error)값을 이용하여 추천기법을 비교 하였다. 실험을 통해 유사도 측정 기법과 예측값 생성 기법의 최적화된 조합을 찾을 수 있었고, 사용자 기반 협업여과와 아이템 기반 협업여과의 성능비교를 통해 아이템 기반 협업여과의 성능이 보다 우수했음을 확인 하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.9
• /
• pp.1905-1912
• /
• 2009

Clustering results based on the FCM algorithm sometimes produces undesirable clustering result through data distribution in the clustered space because data is classified by comparison with membership degree which is calculated by the Euclidean distance between input vectors and clusters. Symmetrical measurement of clusters and fuzzy theory are applied to the classification to tackle this problem. The enhanced FCM algorithm has a low impact with the variation of changing distance about each cluster, middle of cluster and cluster formation. Improved hybrid network of applying FCM algorithm is proposed to classify patterns effectively. The proposed enhanced FCM algorithm is applied to the learning structure between input and middle layers, and normalized delta learning rule is applied in learning stage between middle and output layers in the hybrid network. The proposed algorithms compared with FCM-based RBF network using Max_Min neural network, FMC-based RBF network and HCM-based RBF network to evaluate learning and recognition performances in the two-dimensional coordinated data.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.9 no.2
• /
• pp.96-103
• /
• 2010

This paper studies the performance of STTC with minimum Hamming distance (MHD) mapping in order to improve the bit error rate (BER) performance. Unfortunately, the MHD mapping used in trellis coded modulation (TCM) or multiple trellis coded modulation (MTCM) cannot be directly applied to STTC because the trellis structure of STTC is generally different from that of TCM or MTCM. Therefore, we need a simple modification to apply the MHD mapping concept in STTC. The core of the modification assigns information bits with a Hamming distance in proportion to the sum of the Euclidean distance to trellis branch of STTC. To the best knowledge, this combination has not been considered yet. The BER performance is examined with simulations and the performance of MHD mapping is compared to that of well known natural mapping and Gray mapping on both fast Rayleigh as well as fast Rician fading channels. It is shown that the performance of MHD mapping is much better than that of natural mapping or Gray mapping over fast Rician fading channels, especially.

• Journal of Korean Society of Disaster and Security
• /
• v.9 no.2
• /
• pp.15-21
• /
• 2016

According to the disaster statistics issued by the Ministry of Public Safety and Security, traffic accidents, fire, collapses and others are classified into twenty-three (23) categories. In the past, disasters were mainly caused by the influence of nature, such as typhoon or drought. On the other hand, as society has become city-centered, social disasters' types, frequencies and scales are becoming more diversified and ever-increasing. However, there are no specific criteria and assessment methods that can measure degrees of social disasters-related risks objectively. Therefore, this study targeted traffic accidents, fire and collapses from major social disasters, utilized data that are related to occurrence rate, scale of casualties and scale of property loss in past eight years, and calculated the disaster risk index using the distance (Euclidean distance) between two points on the 3D spatial coordinates, in order to make the objective assessment by social disaster type possible. These results will enable the objective evaluation of risk index of major social disaster to be used as the foundational data when building the national disaster management system.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.22 no.3
• /
• pp.187-194
• /
• 2002

This study was carried out to investigate the ecological characteristics, forage value and similarity among the plant communities of the gazing pasture at Witzenhausen, Germany. Ten plant communities of the different grazing pasture were the Molinio-Arrhenatheretea that was named the class of plant sociological nomenclature. The forage value of the plant communities were ranged from 4.35 to 6.60 grade for roughage qualify. Hemicryptophyte of lift form and mesomorphic of anatomical structure were greately dominated in all the plant communities. The correlation coeffcient between class No. 3 and 4 of plant communities was highest by botanical composition. The clustering analysis by Euclidean distance showed that class No. 9 and 10 of plant communities were closely grouped as affected by the similar botanical composition.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.5
• /
• pp.3434-3439
• /
• 2015

As a criterion of information theoretic learning, the Euclidean distance (ED) of two error probability distribution functions (minimum ED of error, MEDE) has been adopted in nonlinear (decision feedback, DF) supervised equalizer algorithms and has shown significantly improved performance in severe channel distortion and impulsive noise environments. However, the MEDE-DF algorithm has the problem of heavy computational complexity. In this paper, the recursive ED for MEDE-DF algorithm is derived first, and then the feed-forward and feedback section gradients for weight update are estimated recursively. To prove the effectiveness of the recursive gradient estimation for the MEDE-DF algorithm, the number of multiplications are compared and MSE performance in impulsive noise and underwater communication environments is compared through computer simulation. The ratio of the number of multiplications between the proposed DF and the conventional MEDE-DF algorithm is revealed to be $2(9N+4):2(3N^2+3N)$ for the sample size N with the same MSE learning performance in the impulsive noise and underwater channel environment.

• Animal Systematics, Evolution and Diversity
• /
• v.11 no.4
• /
• pp.417-434
• /
• 1995

Three species (0. kinbergi, 0. sarsi and 0 . sarsi vadicola) of the genus Ophiura (Echinodermacta: Ophiuroidea) were sampled from the various localities of Korean waters (Kyonggi Bay in March and September, 1989 and February, 1994; Yellow Sea in September and October, 1992; Southern Sea and Korean Strait in May, 1992; and eastern coasts adjacent to Kangnung in April, July, October, 1993 and January, 1994). Results of the examination of 250 Operational Taxonomical Units (OTUs) are presented based on the 20 morphometric variables to evaluate their taxonomic characters and positions. In cluster analysis, 250 OTUs were divided into 3 phenons (0. kinbergi, 0 . sarsi and 0 . sarsi vadicola) at the Euclidean distance levels of 6.84 and 2 phenons (a phenon composed of 0 . sarsi and 0 . sarsi vadicola and the other of 0. kinbergi) at 7.50. Stepwise discriminant analysis was used in order to produce a good discrimination model and 13 morphological characters (the total number of comb papillae, the number of primary comb papillae and shape of comb papillae (2). etc.) were extracted. The results of canonical discriminant analysis illustrated clear distinction among 3 phenons by the distance of 8.26 between 0 . sarsi and 0. sarsi vadicola, 24.24 between 0 . kinbergi and 0. sarsi vadicola and 21.63 between 0 . kinbergi and 0 . sarsi.