메뉴 건너뛰기
.. 내서재 .. 알림
소속 기관/학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
로그인 회원가입 고객센터 ENG
주제분류

추천
검색
질문

논문 기본 정보

자료유형
학술대회자료
저자정보
Jong-Keuk Lee (동의대학교) Gyeongyong Heo (University of Florida) Soowhan Han (동의대학교) Young Woon Woo (동의대학교)
저널정보
한국멀티미디어학회 한국멀티미디어학회 국제학술대회 MITA 2007
발행연도
2007.8
수록면
161 - 164 (4page)

이용수

표지
📌
연구주제
📖
연구배경
🔬
연구방법
🏆
연구결과
AI에게 요청하기
추천
검색
질문

초록· 키워드

오류제보하기
K-means is one of the simplest unsupervised learning algorithms that solve the clustering problem. However K-means suffers the basic shortcoming: the number of clusters k has to be known in advance. In this paper, we propose an extension of X-means, which is parameter-free and can estimate the number of clusters using Bayesian Information Criterion (BIC). X-means, however, suffers severe over-fitting when the data does not follow a spherical Gaussian distribution due to the diagonal covariance matrix assumption. We introduce two different versions of algorithm; Modified X-means (MX-means) and Generalized X-means (GX-means), which employ one full covariance matrix for one cluster and can estimate the number of clusters efficiently without severe over-fitting and any additional parameters. The algorithms start with one cluster and try to split one cluster iteratively to maximize the BIC score. The former uses the traditional K-means algorithm to find a set of optimal clusters with current k, the number of clusters, which makes it faster than the latter. However it generates incorrectly estimated centers when the clusters are overlapped. The latter uses EM algorithm to estimate the parameters instead of K-means and generates more stable clusters even when the clusters are overlapped. Experiment with synthetic data shows that the proposed methods can provide a robust estimate of the number of clusters and cluster parameters compared to other top-down algorithms.

목차

ABSTRACT
1. INTRODUCTION
2. RELATED RESEARCH
3. MODIFIED X-MEANS
4. GENERALIZED X-MEANS
5. EXPERIMENTAL RESULTS
6. CONCLUSIONS
7. REFERENCES

참고문헌 (0)

참고문헌 신청

함께 읽어보면 좋을 논문

논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!

이 논문의 저자 정보

최근 본 자료

전체보기

댓글(0)

0

UCI(KEPA) : I410-ECN-0101-2012-004-004265342