The Web Science Trust

ATT: Analyzing Temporal Dynamics of Topics and Authors in Social Media

Naveed, Nasir and Sizov, Sergej and Staab, Steffen (2011) ATT: Analyzing Temporal Dynamics of Topics and Authors in Social Media. pp. 1-7. In: Proceedings of the ACM WebSci'11, June 14-17, Koblenz, Germany.

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
277Kb
[img]
Preview
Image (PNG) (Illustration)
35Kb
[img]
Preview
PDF (Presentation) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
1801Kb

Official URL: http://www.websci11.org/fileadmin/websci/Papers/103_paper.pdf

Abstract

Understanding Topical trends and user roles in topic evolution is an important challenge in the field of information retrieval. In this contribution, we present a novel model for analyzing evolution of user’s interests with respect to produced content over time. Our approach Author-Topic-Time model (ATT) addresses this problem by means of Bayesian modeling of relations between authors, latent topics and temporal information. We extend state of the art Latent Dirichlet Allocation (LDA) topic model to incorporate the author and timestamp information for capturing changes in user interest over time with respect to evolving latent topics. We present results of application of the model to the 9 years of scientific publication datasets from CiteSeer showing improved semantically cohesive topic detection and capturing shift in authors interest in relation to topic evolution. We also discuss opportunities of model use in novel mining and recommendation scenarios.

Item Type:Conference or Workshop Item (Paper)
Web Science Comments:WebSci Conference 2011
Subjects:WS2 Artificial Intelligence > WS22 Languages
WS1 Computer Science > WS11 Computability
WS2 Artificial Intelligence > WS24 Basyesian Methods
Web Science Events > Web Science 2011
ID Code:427
Deposited By: Lisa Sugiura
Deposited On:07 Jun 2011 16:53
Last Modified:25 Oct 2011 17:10

Repository Staff Only: item control page

EPrints Logo
Web Science Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.