WCSE 2017
ISBN: 978-981-11-3671-9 DOI: 10.18178/wcse.2017.06.020

Kernel FCM Based on Simulated Annealing Algorithm

Yang Weixin, Tang Lingli, Wang Shengxiang, Wang Chaoliang, Li Chuanrong

Abstract— The time of news topic event is the representation of the temporal characteristics of news, which plays an important role in news information retrieval and mining. According to the current news topic time especially low accuracy of implicit time extraction problem, starting from the characteristics of news analysis, news category feature and web structure mining news topic - time relation model, considering the frequency and position of the theme time, put forward the theme of time extraction algorithm of optimal parameters of the model hierarchical tree model. Randomly selected samples from the Internet for test, the results show that the proposed algorithm has higher accuracy than other methods.

Index Terms— Topic time, Information extraction, Web pages, Hierarchical tree, Explicit time, Implicit time.

Yang Weixin, Tang Lingli, Wang Shengxiang, Wang Chaoliang, Li Chuanrong
Key Laboratory of Quantitative Remote Sensing Information Technology, Chinese Academy of Sciences, CHINA

[Download]


Cite: Yang Weixin, Tang Lingli, Wang Shengxiang, Wang Chaoliang, Li Chuanrong, "Kernel FCM Based on Simulated Annealing Algorithm," Proceedings of 2017 the 7th International Workshop on Computer Science and Engineering, pp. 116-122, Beijing, 25-27 June, 2017.