[Objective] The paper analyzed the feasibility of using Bayesian network for topic tracking, and proposed a new method to improve its performance.[Methods] We constructed two topic tracking models, one with Bayesian Network, and the other with Extended Bayesian Network. The nodes in the models represent terms, events and topics, while the arcs represent relationships among nodes. Finally, we calculated the similarity among topics, events and reports with the Propagation and Evaluation method.[Results] We examined our models on TDT4 data set and found the DET curve of the Bayesian Network model was below the curve of vector space topic model, the former had better performance. The result of extended Bayesian network topic tracking model was 1.7% higher than the first one.[Limitations] Extended Bayesian network topic tracking model was a static topic model while events were generated by the evolution of topics, so the model had limited performance improvement.[Conclusions] The new models can describe the structural relationships among topics, events and stories, and conduct probability inference, which improve the performance of topic tracking effectively.
( Hong Yu, Cang Yu, Yao Jianmin , et al. Descending Kernel Track of Static and Dynamic Topic Models in Topic Tracking[J]. Journal of Software, 2012,23(5):1100-1119.)
[2]
Allan J, Papka R, Lavrenko V . On-Line New Event Detection and Tracking [C]// Proceedings of the 21st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 1998: 37-75.
( Qu Qingtao, Liu Qicheng, Mu Chunxiao . A Parallel Adaptive News Topic Tracking Algorithm Based on N-Gram Language Model[J]. Journal of Shandong University: Engineering Science, 2018,48(6):37-43.)
( Wang Yamin, Hu Yue . Hotspot Detection in Microblog Public Opinion Based on Biterm Topic Model[J]. Journal of Intelligence, 2016,35(11):119-124, 140.)
( Song Lina, Feng Xupeng, Liu Lijun , et al. Microblog Topics Detection Based on SOM Clustering[J]. Application Research of Computers, 2018,35(3):671-674, 679.)
[6]
Xu J M, Wu S F, Hong Y . Topic Tracking with Bayesian Belief Network[J]. Optik, 2014,125(9):2164-2169.
[7]
De Campos L M, Fernández-Luna J M, Huete J F . The BNR Model: Foundations and Performance of a Bayesian Network-Based Retrieval Model[J]. International Journal of Approximate Reasoning, 2003,34(2-3):265-285.
[8]
Doddington G, Fiscus J . The 2002 Topic Detection and Tracking (TDT2002) Task Definition and Evaluation Plan[R]. 2002.
( Zheng Wei, Hou Hongxu, Wu Jing . Application of Bayesian Network for Information Retrieval[J]. Information Science, 2018,36(6):136-141.)
[10]
Turtle H R, Croft W B . Inference Networks for Document Retrieval [C]// Proceedings of the 13th SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 1989: 1-24.
[11]
Ribeiro-Neto B A N, Muntz R . A Belief Network Model for IR [C]// Proceedings of the 19th ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 1996: 253-260.
[12]
Acid S, De Campos L M, Fernández-Luna J M , et al. An Information Retrieval Model Based on Simple Bayesian Networks[J]. International Journal of Intelligent Systems, 2003,18(2):251-265.
( Zhou Nan, Du Pan, Jin Xiaolong , et al. ET-TAG: A Tag Generation Model for the Sub-Topic of Public Opinion Events[J]. Chinese Journal of Computers, 2018,41(7):1490-1503.)
( Zheng Wei, Zhang Yu, Zou Bowei , et al. Research of Chinese Topic Tracking Based on Relevance Model[C]// Proceedings of the 9th China National Conference on Computational Linguistics. Chinese Information Processing Society of China, 2007: 558-563.)