• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.1
• /
• pp.9-16
• /
• 2011

An approach to understand the phase distribution in a multi-phase polycrystalline material is important since it can affect material properties and mechanical behaviors. A proper method is needed to describe the phase distribution. For this purpose, contiguity and probability functions(two-point correlation and lineal-path functions) are investigated for representing the phase distributions of microstructures. The mechanical behaviors are evaluated using the finite element method. The characteristics of probability functions and mechanical reponses of virtual samples are represented. It is confirmed that the topology of phase clustering affects the mechanical behavior of materials and that the strength is reduced as the clustering size increases.

• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.905-922
• /
• 2021

Cluster analysis is one of unsupervised learning techniques used for discovering clusters when there is no prior knowledge of group membership. K-means, one of the commonly used cluster analysis techniques, may fail when the number of variables becomes large. In such high-dimensional cases, it is common to perform tandem analysis, K-means cluster analysis after reducing the number of variables using dimension reduction methods. However, there is no guarantee that the reduced dimension reveals the cluster structure properly. Principal component analysis may mask the structure of clusters, especially when there are large variances for variables that are not related to cluster structure. To overcome this, techniques that perform dimension reduction and cluster analysis simultaneously have been suggested. This study proposes probabilistic reduced K-means, the transition of reduced K-means (De Soete and Caroll, 1994) into a probabilistic framework. Simulation shows that the proposed method performs better than tandem clustering or clustering without any dimension reduction. When the number of the variables is larger than the number of samples in each cluster, probabilistic reduced K-means show better formation of clusters than non-probabilistic reduced K-means. In the application to a real data set, it revealed similar or better cluster structure compared to other methods.

• Journal of KIISE:Software and Applications
• /
• v.27 no.4
• /
• pp.422-431
• /
• 2000

This paper proposes a new codebook generation method, called a PCA-Based VQ, that incorporates the PCA (Principal Component Analysis) technique into VQ (Vector Quantization) codebook design. The PCA technique reduces the data dimensions by transforming input image vectors into the feature vectors. The cluster of feature vectors in the transformed domain is bisectioned into two subclusters by an optimally chosen partitioning hyperplane. We expedite the searching of the optimal partitioning hyperplane that is the most time consuming process by considering that (1) the optimal partitioning hyperplane is perpendicular to the first principal axis of the feature vectors, (2) it is located on the equilibrium point of the left and right cluster's distortions, and (3) the left and right cluster's distortions can be adjusted incrementally. This principal axis bisectioning is successively performed on the cluster whose difference of distortion between before and after bisection is the maximum among the existing clusters until the total distortion of clusters becomes as small as the desired level. Simulation results show that the proposed PCA-based VQ method is promising because its reconstruction performance is as good as that of the SOFM (Self-Organizing Feature Maps) method and its codebook generation is as fast as that of the K-means method.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.29 no.3
• /
• pp.41-63
• /
• 2018

This study aimed at analyzing journals and index terms, authors of the reformed theology and evangelicalism, neutral theological position by using bibliometrical analyzing methods. The analyzing methods are average linkage and neighbor centralities, profile cosine similarities. Especially, when analyzing the relationship between authors, we interpreted the research topic by finding the key shared index terms between the authors. In the journal analysis results, 9 journals were largely clustered together in the two clusters of the reformed theology and evangelicalism, but Presbyterian Theological Quarterly that is thought to be a reformed journal was clustered in evangelical cluster. In the index terms analysis results of the clusters, the reformed theology and evangelicalism were key words representing the two clusters. In the authors' analysis results, we had 9 clusters and the Presbyterian theologian studying the reformed theology had the four clusters and the non-Presbyterian theologian had the 5 clusters. Therefore, we consistently had the two clusters of the reformed theology and evangelicalism in all the analysis of the journals and the index terms, the authors.

• Journal of Korea Water Resources Association
• /
• v.43 no.1
• /
• pp.15-24
• /
• 2010

Regional frequency analysis is often used to overcome the limitation of point frequency analysis to estimate probability rainfall depths. However, point frequency analysis is still used in drought analyses. This study proposed a practical method to categorize the homogeneous regions of drought characteristics for the analyses of regional characteristics of droughts in Korea. Using rainfall data from 58 observation stations managed by the Korea Meteorological Administration, this study calculated drought attributes, i.e., mean drought indices for various durations using the Standardized Precipitation Index (SPI) and drought severities expressed by durations, depth, and intensity. The drought attributes provided useful information for categorizing stations into the hydrological homogeneous regions. This study introduced a cluster analysis with K-means techniques to group observation stations. The cluster analysis grouped observation stations into 6 regions in Korea. The data in the hydrological homogeneous region would be used in spatial analysis of drought characteristics and drought regional frequency analysis.

• Proceeding of EDISON Challenge
• /
• 2014.03a
• /
• pp.591-596
• /
• 2014

본 논문에서는 EDISON_전산열유체 시스템을 활용하여 군집주행 형태 변화에 따른 항력계수 및 연료효율 변화를 분석하였다. 해석 대상은 자동차 형상을 단순화한 모델인 아흐메드 형상(Ahmed body)을 이용하였다. 동종차량 간 거리변화에 따른 항력계수 및 연료효율의 변화, 이종차량의 배열순서변화에 따른 항력계수 및 연료효율을 분석하며 연구를 수행하였다. 연구 결과, 4대의 동종차량 군집주행시 항력계수를 최대 69% 감소할 수 있으며 이에 따라 km 당 0.073L의 연료를 절감할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10a
• /
• pp.581-584
• /
• 2001

신경망이 만들어내는 출력에 대한 정보는 수치적으로 분산되어 신경망에 저장되므로, 인간이 직접 해석하기가 힘들다. 본 논문에서는 LRE(link rule extraction)기법인 NofM 알고리즘의 6단계 중에서 초기 단계인 가중치 군집화 단계를 개선하여 추출되는 규칙들의 전제부에 들어가는 규칙 조건들의 수를 조절함으로써, 추출된 규칙이 입력 특성에 대한 정보를 과잉 일반화하거나, 과잉 구체화하는 것을 피할 수 있음을 실험을 통해 보였다. 일반적으로 NofM 알고리즘에서 가중치들을 군집화한 때는 Join 알고리즘을 사용하는데, 본 논문에서는 Join 알고리즘의 Join condition을 0.05부터 0.25까지 0.05씩 점진적으로 확대하여 클러스터링을 하여줌으로써 신경망의 출력에 중요한 역할을 하는 가중치들을 효과적으로 군집화함을 보였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2007.05a
• /
• pp.1568-1572
• /
• 2007

지역빈도해석은 우리나라와 같이 자료 기간이 짧은 경우 지점빈도해석보다 더 정확한 확률강우량을 산정할 수 있는 기법이다. 지역빈도해석을 통한 확률강우량 산정 결과는 수문학적으로 동질한 지역의 구분 결과에 따라 달라진다. 지역을 구분할 때에는 강우에 영향을 미치는 다양한 변수들이 사용될 수 있다. 변수의 유형과 개수가 지역 구분의 효율성을 좌우하기 때문에 활용 가능한 모든 변수들의 정보를 요약할 수 있는 변수들을 선택하는 것이 지역 구분의 효율성 면에서 유리하다고 할 수 있다. 이런 면에서 지역 구분의 효율성을 증대시킬 목적으로 다변량 분석 기법이 활용될 수 있다. 본 연구에서는 주성분 분석, 요인 분석, Procrustes analysis와 같은 다변량 분석 기법을 활용하여 42개의 강우 관련 변수들을 33개의 변수로 줄일 수 있었다. 분석 결과 변수 개수 감소로 인한 정보 손실은 크지 않은 것으로 나타났다. 따라서 이러한 기법에 의한 변수 차원의 축소는 지역 구분의 효율성 향상에 기여할 수 있는 것으로 판단된다. 선정된 변수들을 바탕으로 군집해석을 수행하여 지역을 구분하였고, L-모멘트에 근거한 이질성척도(H)를 활용하여 구분된 지역의 동질성을 검토하였다. 또한 L-모멘트에 근거한 적합성 척도(Z)를 적용하여 구분된 지역에 적합한 확률분포형을 선정하였고, 선정된 적정 확률분포형을 바탕으로 각 지역에 대한 성장 곡선(growth curve)을 유도하였다.

• Electrical & Electronic Materials
• /
• v.24 no.2
• /
• pp.30-38
• /
• 2011

• Bioindustry News
• /
• v.14 no.4
• /
• pp.29-34
• /
• 2001