Exploring Independent Trends in a Topic-Based Search Engine
J. Perkio, W. Buntine, и S. Perttu. WI'04: IEEE/WIC/ACM International Conference on Web Intelligence, стр. 664--668. (2004)
Аннотация
Topic-based search engines are an alternative to simple keyword search engines that are common in today’s intranets. The temporal behaviour of the topics in a topic model based search engine can be used for trend analysis, which is an important research goal on its own. We apply topic modelling to an online financial newspaper data and show that some of the trends in the topics are consistent with common understanding.
WI'04: IEEE/WIC/ACM International Conference on Web Intelligence
год
2004
страницы
664--668
comment
- use gibbs
- have temporal / trend analysis (although it is not explained in detail)
- have a mechanism to extract "descriptive phrases" for each topic
(pretty topic phrases, pretty labels)
%0 Conference Paper
%1 citeulike:532533
%A Perkio, Jukka
%A Buntine, Wray
%A Perttu, Sami
%B WI'04: IEEE/WIC/ACM International Conference on Web Intelligence
%D 2004
%K engine ir search topic
%P 664--668
%T Exploring Independent Trends in a Topic-Based Search Engine
%U http://cms.dt.uh.edu/Faculty/ChenP/IR/perkio04exploring.pdf
%X Topic-based search engines are an alternative to simple keyword search engines that are common in today’s intranets. The temporal behaviour of the topics in a topic model based search engine can be used for trend analysis, which is an important research goal on its own. We apply topic modelling to an online financial newspaper data and show that some of the trends in the topics are consistent with common understanding.
@inproceedings{citeulike:532533,
abstract = {Topic-based search engines are an alternative to simple keyword search engines that are common in today’s intranets. The temporal behaviour of the topics in a topic model based search engine can be used for trend analysis, which is an important research goal on its own. We apply topic modelling to an online financial newspaper data and show that some of the trends in the topics are consistent with common understanding.},
added-at = {2007-08-15T11:45:58.000+0200},
author = {Perkio, Jukka and Buntine, Wray and Perttu, Sami},
biburl = {https://www.bibsonomy.org/bibtex/260dee28820870efa3af4463f451ac4a8/wnpxrz},
booktitle = {WI'04: IEEE/WIC/ACM International Conference on Web Intelligence},
citeulike-article-id = {532533},
comment = {- use gibbs
- have temporal / trend analysis (although it is not explained in detail)
- have a mechanism to extract "descriptive phrases" for each topic
(pretty topic phrases, pretty labels)},
interhash = {52e529868fe0825af3092b7f3f627c62},
intrahash = {60dee28820870efa3af4463f451ac4a8},
keywords = {engine ir search topic},
pages = {664--668},
priority = {0},
timestamp = {2007-08-15T11:45:58.000+0200},
title = {Exploring Independent Trends in a Topic-Based Search Engine},
url = {http://cms.dt.uh.edu/Faculty/ChenP/IR/perkio04exploring.pdf},
year = 2004
}