Publications
Detailed Information
Functional Data Clustering for High Dimensional Data with Outliers : 이상점이 있는 고차원 데이터의 함수형 군집분석
Cited 0 time in
Web of Science
Cited 0 time in Scopus
- Authors
- Advisor
- 오희석
- Major
- 자연과학대학 통계학과
- Issue Date
- 2014-02
- Publisher
- 서울대학교 대학원
- Keywords
- Functional data clustering ; Robustness ; ES-algorithm ; Pseudo data
- Description
- 학위논문 (석사)-- 서울대학교 대학원 : 통계학과, 2014. 2. 오희석.
- Abstract
- A clustering method for high dimensional data is very difficult because of the curse of dimensionality. In order to reduce the dimension of data, we use the functional data which is obtained by projecting data on a set of basis functions. There are three approaches on functional data clustering: 2-step methods, Nonparametric clustering, Model-based clustering. However, there are few studies on how to cluster the high dimensional data with outliers. In this paper, we suggest new robust functional clustering method using ES-algorithm and k-means clustering.
- Language
- English
- Files in This Item:
- Appears in Collections:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.