Publications
Detailed Information
A Unified Framework of Robust PCA: Use of Robust Unit Approach : 로버스트 유닛을 통한 로버스트 주성분 분석
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 오희석 | - |
dc.contributor.author | 김정음 | - |
dc.date.accessioned | 2017-10-31T08:33:37Z | - |
dc.date.available | 2017-10-31T08:33:37Z | - |
dc.date.issued | 2017-08 | - |
dc.identifier.other | 000000145687 | - |
dc.identifier.uri | https://hdl.handle.net/10371/138090 | - |
dc.description | 학위논문 (석사)-- 서울대학교 대학원 자연과학대학 통계학과, 2017. 8. 오희석. | - |
dc.description.abstract | In this paper, we propose a new framework of robust PCA for improving the robustness and for reflecting various outlier types as well as skewed data. This framework is composed of two concepts: robust unit that is induced by a combination of any PCA procedure and a restriction function as an outlier filter and two-stage strategy that divides and conquers outliers. As a practical implementation of the proposed framework, we develop a robust PCA procedure, termed robust pair PCA (RP-PCA) by coupling a t-distribution-based probabilistic PCA (T-PCA) with our framework. Moreover, for missing data, we suggest a new procedure for handling missing values that fully exploits EM algorithm of T-PCA under the robust unit. Empirical performance of the proposed method is evaluated through numerical studies including simulation study and real data analysis, which demonstrates promising results of the proposed robust method. | - |
dc.description.tableofcontents | 1 Introduction 1
2 Review: T-PCA and ROBPCA 7 2.1. T-PCA 7 2.2. ROBPCA Algorithm 9 3 A New Framework for Robust PCA 11 3.1. Robust Unit and New Framework 11 3.2. ROBPCA Revisited 15 4 Robust Pair PCA 17 4.1. t-Robust Unit 17 4.2. Robust Pair PCA (RP-PCA) 19 5 Simulation Study 20 5.1. Multimodal Data 23 5.2. Unimodal Data 24 5.3. Skewed Data with Skewed Scores and Outliers 25 5.4. Skewed Error Data 27 6 Real Data Analysis 30 6.1. Pendigit Data 30 6.2. Food Data 32 7 RP-PCA with Missing Data 35 8 Conclusion 39 | - |
dc.format | application/pdf | - |
dc.format.extent | 3600604 bytes | - |
dc.format.medium | application/pdf | - |
dc.language.iso | en | - |
dc.publisher | 서울대학교 대학원 | - |
dc.subject | Missing values | - |
dc.subject | Outliers | - |
dc.subject | Principal component analysis | - |
dc.subject | Probabilistic PCA | - |
dc.subject | Robust PCA | - |
dc.subject | Robust unit | - |
dc.subject | Skewed data | - |
dc.subject.ddc | 519.5 | - |
dc.title | A Unified Framework of Robust PCA: Use of Robust Unit Approach | - |
dc.title.alternative | 로버스트 유닛을 통한 로버스트 주성분 분석 | - |
dc.type | Thesis | - |
dc.contributor.AlternativeAuthor | Jungeum Kim | - |
dc.description.degree | Master | - |
dc.contributor.affiliation | 자연과학대학 통계학과 | - |
dc.date.awarded | 2017-08 | - |
- Appears in Collections:
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.