| 한글제목(Korean Title) |
스마트폰 사용자 이동 경로 표현을 위한 계층적 은닉 마르코프 모델 자동 학습 |
| 영문제목(English Title) |
EM Learning of Hierarchical Hidden Markov Model for Smartphone User Dynamic Route Representation |
| 저자(Author) |
바트셀렘
박영택
Batselem Jagvaral
Young Tack Park
|
| 원문수록처(Citation) |
VOL 40 NO. 01 PP. 1482 ~ 1484 (2013. 06) |
한글내용
(Korean Abstract) |
|
영문내용
(English Abstract) |
The model such as the hidden Markov model (HMM) is inefficient in differentiating between signatures of human high level behaviors. The hierarchical hidden Markov model (HHMM) provides more flexible modeling of sequential data than the HMM does by allowing a potentially unbounded number of levels in the hierarchy. The goal of this paper is to study a parameter learning technique to interpret human high level behaviors using the HHMM. We propose a system to recognize human dynamic routes using the HHMM and Expectation Maximization Algorithm (EM). Based on the users recent monthly GPS logdata, the system derives a model that determines the places a person would most likely visit. Also, this study shows how the parameters of the HHMM can be learned using an unsupervised learning techniques such as EM algorithm. |
| 키워드(Keyword) |
|
| 파일첨부 |
PDF 다운로드
|
