A Dynamic Probabilistic Model to Visualise Topic Evolution in Text Streams
A. Kaban, und M. Girolami. Journal of Intelligent Information Systems, 18 (2/3):
107--125(2002)
Zusammenfassung
We propose a novel probabilistic method, based on latent variable models, for unsupervised topographic visualisation of dynamically evolving, coherent textual information. This can be seen as a complementary tool for topic detection and tracking applications. This is achieved by the exploitation of the a priori domain knowledge available, that there are relatively homogeneous temporal segments in the data stream. In a different manner from topographical techniques previously utilized for static text collections, the topography is an outcome of the coherence in time of the data stream in the proposed model. Simulation results on both toy-data settings and an actual application on Internet chat line discussion analysis is presented by way of demonstration.
%0 Journal Article
%1 citeulike:393563
%A Kaban, Ata
%A Girolami, Mark
%D 2002
%J Journal of Intelligent Information Systems
%K socialnets
%N 2/3
%P 107--125
%T A Dynamic Probabilistic Model to Visualise Topic Evolution in Text Streams
%U http://www.wkap.nl/article.pdf?391242
%V 18
%X We propose a novel probabilistic method, based on latent variable models, for unsupervised topographic visualisation of dynamically evolving, coherent textual information. This can be seen as a complementary tool for topic detection and tracking applications. This is achieved by the exploitation of the a priori domain knowledge available, that there are relatively homogeneous temporal segments in the data stream. In a different manner from topographical techniques previously utilized for static text collections, the topography is an outcome of the coherence in time of the data stream in the proposed model. Simulation results on both toy-data settings and an actual application on Internet chat line discussion analysis is presented by way of demonstration.
@article{citeulike:393563,
abstract = {We propose a novel probabilistic method, based on latent variable models, for unsupervised topographic visualisation of dynamically evolving, coherent textual information. This can be seen as a complementary tool for topic detection and tracking applications. This is achieved by the exploitation of the a priori domain knowledge available, that there are relatively homogeneous temporal segments in the data stream. In a different manner from topographical techniques previously utilized for static text collections, the topography is an outcome of the coherence in time of the data stream in the proposed model. Simulation results on both toy-data settings and an actual application on Internet chat line discussion analysis is presented by way of demonstration.},
added-at = {2006-06-16T10:34:37.000+0200},
author = {Kaban, Ata and Girolami, Mark},
biburl = {https://www.bibsonomy.org/bibtex/2f81bc9315be5c3fb426ee22b2508c4d6/ldietz},
citeulike-article-id = {393563},
comment = {Special Issue on Automated Text Categorization},
interhash = {4312e35ac215d21937a017718a4bf3df},
intrahash = {f81bc9315be5c3fb426ee22b2508c4d6},
journal = {Journal of Intelligent Information Systems},
keywords = {socialnets},
number = {2/3},
pages = {107--125},
priority = {3},
timestamp = {2006-06-16T10:34:37.000+0200},
title = {A Dynamic Probabilistic Model to Visualise Topic Evolution in Text Streams},
url = {http://www.wkap.nl/article.pdf?391242},
volume = 18,
year = 2002
}