Topic tracking with Bayesian belief network. Xu, J., Wu, S., & Hong, Y. Optik - International Journal for Light and Electron Optics, 125(9):2164-2169, 5, 2014.
Topic tracking with Bayesian belief network [link]Website  doi  abstract   bibtex   
The task of topic tracking is to monitor a stream of stories and find all subsequent stories that discuss the same topic. Using Bayesian belief network we give three topic tracking models: a static topic model BSTM and two dynamic topic models BDTM-I, BDTM-II. BDTM-II merges the advantages of BSTM and BDTM-I, has better tracking performance than the former two, and effectively alleviates topic drift phenomenon. Applying unrelated coming stories to update BDTM-I and BDTM-II can filter noises existed in topics. Experiments on TDT corpora show that BSTM decreases (Cdet)norm by 5.5% comparing to VSM, BDTM-II decreases (Cdet)norm by 6.3% and 6.0% comparing to BSTM and BDTM-I respectively, using unrelated stories can improve the tracking performance.
@article{
 title = {Topic tracking with Bayesian belief network},
 type = {article},
 year = {2014},
 keywords = {Bayesian belief network,Dynamic topic model,Static topic model,Topic tracking},
 pages = {2164-2169},
 volume = {125},
 websites = {http://www.sciencedirect.com/science/article/pii/S0030402613013909},
 month = {5},
 id = {8aa54219-cf9f-366e-96d2-10eca0e5ac0c},
 created = {2015-04-11T18:56:31.000Z},
 accessed = {2015-04-11},
 file_attached = {false},
 profile_id = {95e10851-cdf3-31de-9f82-1ab629e601b0},
 group_id = {09500bf6-14e8-379d-a953-ea715d61ca19},
 last_modified = {2017-03-14T14:28:50.201Z},
 read = {false},
 starred = {false},
 authored = {false},
 confirmed = {true},
 hidden = {false},
 private_publication = {false},
 abstract = {The task of topic tracking is to monitor a stream of stories and find all subsequent stories that discuss the same topic. Using Bayesian belief network we give three topic tracking models: a static topic model BSTM and two dynamic topic models BDTM-I, BDTM-II. BDTM-II merges the advantages of BSTM and BDTM-I, has better tracking performance than the former two, and effectively alleviates topic drift phenomenon. Applying unrelated coming stories to update BDTM-I and BDTM-II can filter noises existed in topics. Experiments on TDT corpora show that BSTM decreases (Cdet)norm by 5.5% comparing to VSM, BDTM-II decreases (Cdet)norm by 6.3% and 6.0% comparing to BSTM and BDTM-I respectively, using unrelated stories can improve the tracking performance.},
 bibtype = {article},
 author = {Xu, Jian-min and Wu, Shu-fang and Hong, Yu},
 doi = {10.1016/j.ijleo.2013.10.044},
 journal = {Optik - International Journal for Light and Electron Optics},
 number = {9}
}

Downloads: 0