• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 1999년도 추계학술대회 논문집 - 한국공작기계학회
• /
• pp.345-350
• /
• 1999

The effectiveness of software error compensation for thermally induced machine tool errors relies on the prediction accuracy of the pre-established thermal error models. The selection of optimal sensor locations is the most important in establishing these empirical models. In this paper, a methodology for the selection of optimal sensor locations is proposed to establish a robust linear model which is not subjected to collinearity. Correlation coefficient and time delay are used as thermal parameters for optimal sensor location. Firstly, thermal deformation and temperatures are measured with machine tools being excited by sinusoidal heat input. And then, after correlation coefficient and time delays are calculated from the measured data, the optimal sensor location is selected through hard c-means clustering and sequential selection method. The validity of the proposed methodology is verified through the estimation of thermal expansion along Z-axis by spindle rotation.

• 정보처리학회논문지B
• /
• 제10B권3호
• /
• pp.311-320
• /
• 2003

고유의 텍스쳐 성분에만 최적 반응을 하는 최적 필터(optimal filter)는 다중 텍스쳐 영상으로부터 원하는 텍스쳐 성분을 추출하기 위한 가장 뛰어난 기술이다. 그러나 기존의 최적필터 설계 방법들은 영상에 내재된 텍스쳐 정보가 사전에 주어지는 교사적 방법이 대부분이며, 내재된 텍스쳐 인식을 기반으로 하는 완전 비교사적인 방법에 관한 연구는 거의 이루어지고 있지 않은 실정이다. 따라서 본 논문에서는 효율적인 텍스쳐 분석을 위한 비교사 학습 방법과 가버필터의 주파수 대역 통과형 특징을 이용한 새로운 최적 필터 설계 방법을 제안한다. 제안한 방법은 자기조직형 신경회로망에 의해서 영상에 내재된 텍스쳐 영역을 블록 단위로 군화(clustering)하며, 가버필터의 최적 주파수는 인식된 텍스쳐 오브젝트(texture objects)의 공간 주파수를 분석한 최적 주파수에 동조(turning)한다. 그리고 설계된 최적 가버필터의 성능 평가를 위해서는 다양한 형태의 다중 텍스쳐 영상을 생성하여 내재된 텍스쳐 오브젝트를 추출함으로써 성공적인 결과를 보인다.

• 대한토목학회논문집
• /
• 제29권3D호
• /
• pp.347-356
• /
• 2009

본 연구는 기존의 도로기능분류 정의와 방법론을 벗어나 교통특성에 따른 도로분류 방법론인 도로특성분류를 기초로 분석을 수행하였다. 도로특성분류에 대한 일련의 과정 중에서 다양한 교통특성을 반영하는 설명변수를 기초로 요인점수를 산출하고, 동질한 도로구간을 그룹핑하는 군집화 분석과정과 적정 군집수 도출에 따른 군집결과비교에 본 연구는 초점을 맞추었다. 도로분류를 위해 병합적 계층 군집분석인 Ward법, 비계층적 군집분석인 K-means법, 자율신경 회로망을 이용한 K-SOM을 사용하여 비교분석하였다. 각 군집기법에 대한 결과를 토대로 비교분석한 결과, 군집 수 5 이하에서는 K-means법, 군집 수 14 이상에서는 Kohonen selforganizing maps가 가장 우수한 것으로 나타났으며, 군집수 5~9사이에서는 Ward법과 Kmeans법의 군집 성능이 불규칙한 패턴을 보임에 따라 세밀한 결과분석을 통해 우수성을 결정하는 것이 바람직할 것으로 분석되었다. 본 연구결과는 다양한 교통특성을 고려한 도로구간의 군집 속성을 분석하고 예측하는 분류화 작업에 중요한 기초적인 자료로 사용될 것으로 기대된다.

• 산업경영시스템학회지
• /
• 제40권2호
• /
• pp.92-98
• /
• 2017

Data clustering determines a group of patterns using similarity measure in a dataset and is one of the most important and difficult technique in data mining. Clustering can be formally considered as a particular kind of NP-hard grouping problem. K-means algorithm which is popular and efficient, is sensitive for initialization and has the possibility to be stuck in local optimum because of hill climbing clustering method. This method is also not computationally feasible in practice, especially for large datasets and large number of clusters. Therefore, we need a robust and efficient clustering algorithm to find the global optimum (not local optimum) especially when much data is collected from many IoT (Internet of Things) devices in these days. The objective of this paper is to propose new Hybrid Simulated Annealing (HSA) which is combined simulated annealing with K-means for non-hierarchical clustering of big data. Simulated annealing (SA) is useful for diversified search in large search space and K-means is useful for converged search in predetermined search space. Our proposed method can balance the intensification and diversification to find the global optimal solution in big data clustering. The performance of HSA is validated using Iris, Wine, Glass, and Vowel UCI machine learning repository datasets comparing to previous studies by experiment and analysis. Our proposed KSAK (K-means+SA+K-means) and SAK (SA+K-means) are better than KSA(K-means+SA), SA, and K-means in our simulations. Our method has significantly improved accuracy and efficiency to find the global optimal data clustering solution for complex, real time, and costly data mining process.

• 품질경영학회지
• /
• 제18권1호
• /
• pp.48-54
• /
• 1990

The purpose of this study is to design effective working systems which adapt to change in human needs by developing an method which forms into optimal groups using the measure of cohesion. Two main results can be derived from the study as follows : First, the clustering method based on the entropic measure of cohesion is predominant with respect to any other methods proposed in designing the work groups, since this clustering criterion includes symmetrical relations of total work groups and the dissimilarity as well as the similarity relations of predicate value, the clustering method based on this criterion is suitable for designing the new work structure. Second, total work group is clustered as the workers who have the equal predicate value and then clustering results are produced through the combinatorial agglomerative hierarchical clustering method. This clustering method present more economic results than the method that clustering the total work group do.

• Communications for Statistical Applications and Methods
• /
• 제25권3호
• /
• pp.263-274
• /
• 2018

One of the common issues in large dataset analyses is to detect and construct homogeneous groups of objects in those datasets. This is typically done by some form of clustering technique. In this study, we present a divisive hierarchical clustering method for two monothetic characteristics of histogram data. Unlike classical data points, a histogram has internal variation of itself as well as location information. However, to find the optimal bipartition, existing divisive monothetic clustering methods for histogram data consider only location information as a monothetic characteristic and they cannot distinguish histograms with the same location but different internal variations. Thus, a divisive clustering method considering both location and internal variation of histograms is proposed in this study. The method has an advantage in interpreting clustering outcomes by providing binary questions for each split. The proposed clustering method is verified through a simulation study and applied to a large U.S. house property value dataset.

• 전자공학회논문지CI
• /
• 제46권4호
• /
• pp.79-86
• /
• 2009

본 논문에서는 노드 밀도를 고려하여 클러스터를 형성함으로써 데이터 병합 효과를 최대화하고, 에너지 소모를 줄일 수 있는 새로운 클러스터링 기법을 제안한다. 이 기법은 최적의 데이터 병합율을 보장할 수 있도록 클러스터 크기를 결정하기 때문에 메시지 전송 반경을 줄이고, 클러스터간의 간섭을 최소화할 수 있다. 또한 지역적으로 인접한 다수의 노드들을 클러스터로 구성하고 멤버 노드로부터 수신된 데이터를 병합하여 전송함으로써 에너지 소모를 줄인다. 시뮬레이션을 통하여 센서 네트워크를 구성하는 노드의 에너지 소모를 최소화하면 센서 네트워크의 생존시간을 연장할 수 있음을 검증한다. 또한 제안한 클러스터링 기법이 기존의 LEACH 클러스터링 기법에 비해 성능이 우수함을 보인다.

• Communications for Statistical Applications and Methods
• /
• 제6권1호
• /
• pp.53-68
• /
• 1999

To improve the slow convergence property of the steepest ascent type algorithm for continuous D-optimal design problems. we develop a new algorithm. We apply the nonlinear system of equations as the necessary condition of optimality and develop the two-point algorithm that solves the problem of clustering. Because of the nature of the steepest coordinate ascent algorithm avoiding the problem of clustering itself helps the improvement of convergence speed. The numerical examples show the performances of the new method is better than those of various steepest ascent algorithms.

• 한국농공학회지
• /
• 제45권3호
• /
• pp.73-83
• /
• 2003

Methods of fuzzy C-means have been used to characterize geotechnical information from static cone penetration data. As contrary with traditional classification methods such as Robertson classification chart, the FCM expresses classes not conclusiveness but fuzzy. The results show that the FCM is useful to characterize ground information that can not be easily found by using normal classification chart. But optimal number of classes may not be easily defined. So, the optimal number of classes should be determined considering not only technical measures but engineering aspects.

• Communications for Statistical Applications and Methods
• /
• 제11권2호
• /
• pp.369-380
• /
• 2004

Clustering algorithms with different similarity measures are commonly used to find an optimal clustering or close to original clustering. The recovery level of using Euclidean distance and distances transformed from correlation coefficients is evaluated and compared using Rand's (1971) C statistic. The C values present how the resultant clustering is close to the original clustering. In simulation study, the recovery level is improved by applying the correlation coefficients between objects. Using the data set from Spellman et al. (1998), the recovery levels with different similarity measures are also presented. In general, the recovery level of true clusters was increased by using the correlation coefficients.