결측치를 포함한 데이터의 k-평균 군집분석 방법 비교 = Comparison of k-mean clustering with missing data

다국어 초록 (Multilingual Abstract)

Cluster analysis is an unsupervised learning method to find heterogeneous clusters that capture similarities among items and separate different items into different clusters. Various cluster analysis techniques have been proposed, and the k-means clustering method, which minimizes the sum of Euclidean distances between cluster centroids and individual entities, is widely recognized as a standard cluster analysis method. When data include missing values, it is challenging to conduct cluster analysis, because it is impossible to calculate distances between centroids of clusters and incomplete items, resulting in excluding classification of these items. Techniques have been suggested to handle missing values in k-means clustering, including conducting cluster analysis after imputation of missing values or cluster analysis based on available information. In this study, we explore methods to perform k-means cluster analysis on data with missing values and evaluate performance of these methods using a simulation. The results of simulation studies indicate that conducting k-means cluster analysis after imputation yields the better performance than the one based on available information. Among the various imputation methods, k-nearest neighbors imputation performed the best.

참고문헌 (Reference)

1 윤혜경 ; 김태영 ; 김용하, "군집분석을 활용한 포스트휴먼 교양의 범주화에 관한 연구" 한국자료분석학회 23 (23): 209-217, 2021

2 송주원, "결측자료의 k-평균 군집분석" 한국자료분석학회 19 (19): 689-697, 2017

3 송주원, "결측자료 분석에서 결측 비율이 결측자료 k-평균 군집분석에 미치는 영향" 한국자료분석학회 19 (19): 1273-1282, 2017

4 Van Buuren, S., "mice : Multivariate imputation by chained equations in R" 45 : 1-67, 2011

5 Chi, J. T., "k-pod : A method for k-means clustering of missing data" 70 (70): 91-99, 2016

6 Hastie T, "impute: impute: Imputation for microarray data. R packageversion 1.74.1"

7 Little, R. J. A., "Statistical Analysis With Missing Data" John Wiley 2002

8 Doove, L. L., "Recursive partitioning for missing data imputation in thepresence of interaction effects" 72 : 92-104, 2014

9 이은진 ; 김영서 ; 홍세희, "Network and Cluster Analysis of the Funding Relationship among the U.N. Agencies and State Donors" 한국자료분석학회 23 (23): 2547-2562, 2021

10 Rubin, D. B., "Multiple Imputation for Nonresponse in Surveys" John Wiley 1987

11 Hunt, L., "Mixture model clustering for mixed data with missing information" 41 (41): 429-440, 2003

12 Little, R. J., "Missing-data adjustments in large surveys" 6 (6): 287-296, 1988

13 Du, M., "Grid-Based Clustering Using Boundary Detection" 24 (24): 2022

14 Hubert, L., "Comparing partitions" 2 : 193-218, 1985

15 Wagstaff, K., "Clustering with missing values: No imputation required" Illinois Institute of Technology 649-658, 2004

16 Strehl, A., "Cluster ensembles-a knowledge reuse framework for combining multiple partitions" 3 : 583-617, 2002

17 Pfaffel, O, "ClustImpute: An R package for k-means clustering with build-in missing data imputation, Rpackage version 0.2.4"

18 Moorthy, KA, "A review on missing value imputation algorithms for microarraygene expression data" 9 : 18-22, 2014

19 Dunn, J. C., "A Fuzzy Relative of the ISODATA Process and Its Use in Detecting Compact Well-SeparatedClusters" 3 (3): 32-57, 1973