dc.citation.conferencePlace |
PO |
- |
dc.citation.conferencePlace |
Lisbon |
- |
dc.citation.endPage |
1619 |
- |
dc.citation.startPage |
1610 |
- |
dc.citation.title |
Empirical Methods in Natural Language Processing |
- |
dc.contributor.author |
Kim, Taehoon |
- |
dc.contributor.author |
Choi, Jaesik |
- |
dc.date.accessioned |
2023-12-19T22:06:20Z |
- |
dc.date.available |
2023-12-19T22:06:20Z |
- |
dc.date.created |
2015-09-07 |
- |
dc.date.issued |
2015-09-20 |
- |
dc.description.abstract |
Modeling non-stationary time-series data for making predictions is a challenging but important task. One of the key issues is to identify long-term changes accurately in time-varying data. Bayesian Online Change Point Detection (BO-CPD) algorithms efficiently detect long-term changes without assuming the Markov property which is vulnerable to local signal noise. We propose a Document based BO-CPD (DBO-CPD) model which automatically detects long-term temporal changes of continuous variables based on a novel dynamic Bayesian analysis which combines a non-parametric regression, the Gaussian Process (GP), with generative models of texts such as news articles and posts on social networks. Since texts often include important clues of signal changes, DBO-CPD enables the accurate prediction of long-term changes accurately. We show that our algorithm outperforms existing BO-CPDs in two real-world datasets:stock prices and movie revenues. |
- |
dc.identifier.bibliographicCitation |
Empirical Methods in Natural Language Processing, pp.1610 - 1619 |
- |
dc.identifier.scopusid |
2-s2.0-84959884669 |
- |
dc.identifier.uri |
https://scholarworks.unist.ac.kr/handle/201301/32822 |
- |
dc.identifier.url |
http://aclweb.org/anthology/D15-1184 |
- |
dc.language |
영어 |
- |
dc.publisher |
Conference on Empirical Methods in Natural Language Processing, EMNLP 2015 |
- |
dc.title |
Reading Documents for Bayesian Online Change Point Detection |
- |
dc.type |
Conference Paper |
- |
dc.date.conferenceDate |
2015-09-17 |
- |