DSpace university logo mark
Advanced Search
Japanese | English 

NAOSITE : Nagasaki University's Academic Output SITE > Faculty of Engineering > Articles in academic journal >

Bag of Timestamps: A Simple and Efficient Bayesian Chronological Mining


File Description SizeFormat
LNCS5446_556.pdf269.89 kBAdobe PDFView/Open

Title: Bag of Timestamps: A Simple and Efficient Bayesian Chronological Mining
Authors: Masada, Tomonari / Takasu, Atsuhiro / Hamada, Tsuyoshi / Shibata, Yuichiro / Oguri, Kiyoshi
Issue Date: May-2009
Publisher: Springer Berlin
Citation: Lecture Notes in Computer Science, 5446, pp.556-561; 2009
Abstract: In this paper, we propose a new probabilistic model, Bag of Timestamps (BoT), for chronological text mining. BoT is an extension of latent Dirichlet allocation (LDA), and has two remarkable features when compared with a previously proposed Topics over Time (ToT), which is also an extension of LDA. First, we can avoid overfitting to temporal data, because temporal data are modeled in a Bayesian manner similar to word frequencies. Second, BoT has a conditional probability where no functions requiring time-consuming computations appear. The experiments using newswire documents show that BoT achieves more moderate fitting to temporal data in shorter execution time than ToT.
Description: Advances in Data and Web Management. Joint International Conferences, APWeb/WAIM 2009 Suzhou, China, April 2-4, 2009 Proceedings
URI: http://hdl.handle.net/10069/22149
ISBN: 978-3-642-00671-5
ISSN: 03029743
DOI: 10.1007/978-3-642-00672-2
Rights: (c) Springer-Verlag Berlin Heidelberg 2009 / The original publication is available at www.springerlink.com
Type: Journal Article
Text Version: author
Appears in Collections:Articles in academic journal

Citable URI : http://hdl.handle.net/10069/22149

All items in NAOSITE are protected by copyright, with all rights reserved.

 

Valid XHTML 1.0! Copyright © 2006-2015 Nagasaki University Library - Feedback Powerd by DSpace