• 한국콘텐츠학회논문지
• /
• 제6권1호
• /
• pp.23-30
• /
• 2006

Temporal 데이터의 클러스터링 방법론 중의 하나로 모델기반 방법론이 있다. 이는 각 클러스터에 대하여 오토마타기반의 모델을 가정하는 것이다. 개별 모델을 추출하기 위해서는 먼저 전체 데이터에 대한 적합한 모델을 찾는 것이 필요하다. 전체에 대한 모델은 데이터집합에 대한 최적의 클러스터의 수를 결정함으로 개별 모델 구축의 준비를 완료한다. 본 연구에서는 클러스터 수를 결정하기 위한 기준인 베이지안 정보기준(BIC : Bayesian Information Criterion) 근사법의 활용도를 검증하고 데이터 크기와 BIC 값의 상관관계를 파악함으로 탐색 효율을 높이는 방안을 제안한다. 실험에서는 인위적 모델을 통하여 생성된 인공적인 여러 형태의 데이터집합을 활용하여 BIC근사 측도의 활용성에 대해 살펴보았다. 실험결과에서 보여주는 것처럼 BIC 근사 측도는 데이터의 크기가 비교적 클 경우에 올바른 파티션의 사이즈를 추정함을 확인하였다.

• 천문학회지
• /
• 제27권1호
• /
• pp.61-76
• /
• 1994

We investigate the dynamical evolution of globular clusters under the diffusion, the Galactic tide, and the presence of halo black holes. We compare the results with our previous work which considers the diffusion processes and the Galactic tide. We find the followings: (1) The black holes contribute the expansion of the outer part of the cluster. (2) There is no evidence for dependence on the orbital phase of the cluster as in our previous work. (3) The models of linear and Gaussian velocity distribution for the halo black holes do not show any significant differences in all cases. (4) The perturbation of black holes reduces the number of stars in lower energy regions. (5) There is a significant number of stars with retrograde orbits beyond the cutoff radius especially in the case of diffusion and the perturbation of black holes.

• 천문학회지
• /
• 제37권5호
• /
• pp.433-438
• /
• 2004

Clusters of galaxies generally form by the gravitational merger of smaller clusters and groups. Major cluster mergers are the most energetic events in the Universe since the Big Bang. The basic properties of cluster mergers and their effects are discussed. Mergers drive shocks into the intracluster gas, and these shocks heat the intracluster gas. As a result of the impulsive heating and compression associated with mergers, there is a large transient increase in the X-ray luminosities and temperatures of merging clusters. These merger boost can affect X-ray surveys of clusters and their cosmological interpretation. Similar boosts occur in the strong lensing cross-sections and Sunyaev-Zeldovich effect in merging clusters. Merger shock and turbulence associated with mergers should also (re)accelerate nonthermal relativistic particles. As a result of particle acceleration in shocks and turbulent acceleration following mergers, clusters of galaxies should contain very large populations of relativistic electrons and ions. Observations and models for the radio, extreme ultraviolet, hard X-ray, and gamma-ray emission from nonthermal particles accelerated in these shocks will also be described. Gamma-ray observations with GLAST seem particularly promising.

• 천문학회지
• /
• 제35권2호
• /
• pp.75-85
• /
• 2002

We present N-body simulations of globular clusters including gravitational field of the Galaxy, in order to study effects of tidal field systematically on the shape of outer parts of globular clusters using NBODY6. The Galaxy is assumed to be composed of central bulge and outer halo. We mvestigate the cluster of multi-mass models with a power-law initial mass function (IMF) starting with different initial masses, initial number of particles, different slopes of the IMF and different orbits of the cluster. We have examined the general evolution of the clusters, the shape of outer parts of the clusters, density profiles and the direction of tidal tails. The density profiles appear to become somewhat shallower just outside the tidal boundary consistent with some observed data. The position angle of the tidal tall depends on the location in the Galaxy as well as the direction of the motion of. clusters. We found that the clusters become more elongated at the apogalacticon than at the pengalacticon. The tidal tails may be used to trace the orbital paths of globular clusters.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제6권2호
• /
• pp.138-143
• /
• 2006

In this paper, we have presented a Sequential Agglomerative Hierarchical Nested (SAHN) algorithm-based data clustering method in fuzzy inference system to achieve optimal performance of fuzzy model. SAHN-based algorithm is used to give possible range of number of clusters with cluster centers for the system identification. The axes of membership functions of this fuzzy model are optimized by using cluster centers obtained from clustering method and the consequence parameters of the fuzzy model are identified by standard least square method. Finally, in this paper, we have observed our model's output performance using the Box and Jenkins's gas furnace data and Sugeno's non-linear process data.

• Asia pacific journal of information systems
• /
• 제28권4호
• /
• pp.308-319
• /
• 2018

This paper aims at providing valuable insights on Financial Fraud Detection on a mobile money transactional activity. We have predicted and classified the transaction as normal or fraud with a small sample and massive data set using Azure and Spark ML, which are traditional systems and Big Data respectively. Experimenting with sample dataset in Azure, we found that the Decision Forest model is the most accurate to proceed in terms of the recall value. For the massive data set using Spark ML, it is found that the Random Forest classifier algorithm of the classification model proves to be the best algorithm. It is presented that the Spark cluster gets much faster to build and evaluate models as adding more servers to the cluster with the same accuracy, which proves that the large scale data set can be predictable using Big Data platform. Finally, we reached a recall score with 0.73, which implies a satisfying prediction quality in predicting fraudulent transactions.

• 한국CDE학회논문집
• /
• 제13권1호
• /
• pp.1-7
• /
• 2008

Data compression becomes increasingly an important issue for reducing data storage spaces as well as transmis-sion time in network environments. In 3D geometric models, the normal vectors of faces or meshes take a major portion of the data so that the compression of the vectors, which involves the trade off between the distortion of the images and compression ratios, plays a key role in reducing the size of the models. So, raising the compression ratio when the normal vector is compressed and minimizing the visual distortion of shape model's shading after compression are important. According to the recent papers, normal vector compression is useful to heighten com-pression ratio and to improve memory efficiency. But, the study about distortion of shading when the normal vector is compressed is rare relatively. In this paper, new normal vector compression method which is clustering normal vectors and assigning Representative Normal Vector (RNV) to each cluster and using the angular deviation from actual normal vector is proposed. And, using this new method, Visually Undistinguishable Lossy Compression (VULC) algorithm which distortion of shape model's shading by angular deviation of normal vector cannot be identified visually has been developed. And, being applied to the complicated shape models, this algorithm gave a good effectiveness.

• Genomics & Informatics
• /
• 제20권3호
• /
• pp.34.1-34.11
• /
• 2022

Multilevel analysis is an appropriate and powerful tool for analyzing hierarchical structure data widely applied from public health to genomic data. In practice, however, we may lose the information on multiple nesting levels in the multilevel analysis since data may fail to capture all levels of hierarchy, or the top or intermediate levels of hierarchy are ignored in the analysis. In this study, we consider a multilevel linear mixed effect model (LMM) with single imputation that can involve all data hierarchy levels in the presence of missing top or intermediate-level clusters. We evaluate and compare the performance of a multilevel LMM with single imputation with other models ignoring the data hierarchy or missing intermediate-level clusters. To this end, we applied a multilevel LMM with single imputation and other models to hierarchically structured cohort data with some intermediate levels missing and to simulated data with various cluster sizes and missing rates of intermediate-level clusters. A thorough simulation study demonstrated that an LMM with single imputation estimates fixed coefficients and variance components of a multilevel model more accurately than other models ignoring data hierarchy or missing clusters in terms of mean squared error and coverage probability. In particular, when models ignoring data hierarchy or missing clusters were applied, the variance components of random effects were overestimated. We observed similar results from the analysis of hierarchically structured cohort data.

• 한국항해항만학회지
• /
• 제30권1호
• /
• pp.35-43
• /
• 2006

본 연구는 교통안전공단과 사단법인 녹색교통운동이 공동으로 개발한 교통문화지수와 관련한 2002년과 2003년의 전국 81개 도시 자료를 토대로 통계적 분석을 행하여 이들 대상도시들을 유형화하고, 집단별 영향요인에 근거하여 교통사고 예방대책들을 제시하고자 하였다. 먼저 교통문화지수와 영향요인들에 대한 주성분분석 결과로는 4개의 주성분으로 구분 지울 수 있었으며, 도시 특성별 최적 집단 수는 4개가 적합한 것으로 나타났다. 또한 이들 유형화된 집단별 교통문화지수에의 영향요인을 단계별 다중 회귀분석법을 이용하여 분석한 결과, 4개 집단 모두 높은 설명력을 갖는 회귀모형을 구축할 수 있었다. 이에 따라 각 집단별 교통사고 예방대책들을 구체적으로 제시할 수 있었으며, 아울러 투자된 시설이 얼마나 교통사고 예방에 효과적이었는가를 분석할 필요성이 있음을 향후의 연구 과제로 제시하였다.

• 융합보안논문지
• /
• 제16권7호
• /
• pp.21-29
• /
• 2016

현재 우리사회는 네트워크를 통하여 실시간으로 교류되는 다양한 정보 환경에 노출되어 있다. 특히 정부의 의료정책은 대국민의료서비스 질을 향상시키기 위해 원격진료의 시행을 서두르고 있다. 이러한 원격진료의 시행은 향후 지역에 상관없이 맞춤형 환자 진료를 위한 빅 데이터 기반의 진료 정보 구축도 함께 요구하고 있다. 본 논문은 빅 데이터 기반의 권역별 의료클러스터 구축과 이에 대한 서비스 가용성을 해치는 공격이 발생할 경우 해당 공격을 탐지하고 적절한 대응이 가능한 방어 및 보안 협력모델을 제안하고 있다. 이를 위하여 동일 병원정보시스템으로 전국에 고루 분포된 지방의료원을 권역별 가상 의료클러스터 본부로 하는 네트워크 구성을 제안하였다. 아울러 의료클러스터에 발생할 수 있는 IP Spoofing 공격과 이에 따른 DDoS 공격에 실시간으로 대응 가능한 상호협력 보안 모델을 설계하여 단일 체계, 단일 보안정책이 가지는 한계성도 극복할 수 있도록 하였다.