文献类型：期刊文献

英文题名：Rumor events detection enhanced by encoding sentimental information into time series division and word representations

作者：Wang, Zhihong[1];Guo, Yi[1,2,3]

机构：[1]East China Univ Sci & Technol, Shanghai, Peoples R China;[2]Natl Engn Lab Big Data Distribut & Exchange Techn, Shanghai, Peoples R China;[3]Shanghai Engn Res Ctr Big Data & Internet Audienc, Shanghai, Peoples R China

年份：2020

卷号：397

起止页码：224

外文期刊名：NEUROCOMPUTING

收录：;EI(收录号：20201008251940);WOS:【SCI-EXPANDED(收录号:WOS:000535918300006)】；

基金：This research is financially supported by The National Key Research and Development Program of China (grant number 2018YFC0807105), National Natural Science Foundation of China (grant number 61462073) and Science and Technology Committee of Shanghai Municipality (STCSM) (under grant numbers 17DZ1101003, 18511106602 and 18DZ2252300).

语种：英文

外文关键词：Rumor events detection; Sentiment dictionary; Dynamic time series; Cascaded gated recurrent unit; Online social networks

摘要：Online Social Networks (OSNs) is an ideal place for spreading rumor events as it is convenient in information production and dissemination. Automatically debunking these rumor events is important to pursue and restore the truth. However, it is a challenging task to employ traditional classification approaches for rumor events detection since they rely on hand-crafted features that require daunting manual efforts. Besides, we observe that the various posts of each rumor event will debate its realness over time. Different individuals also have different emotional reactions to events, which will affect others' identification. Thus, this paper firstly employs an automatic construction method to develop a Sentiment Dictionary (SD) to capture the fine-grained human emotional reactions to different events. Secondly, a Two-steps Dynamic Time Series (TsDTS) algorithm, involving the sentimental information in the division process, is elaborated to retain the time-span distribution information of microblog events in a natural manner. At last, a novel two-layer Cascaded Gated Recurrent Unit (CGRU) model based on the SD and the TsDTS algorithm is proposed for rumor events detection, named as SD-TsDTS-CGRU. Experimental results on real datasets from OSNs demonstrate that our proposed SD-TsDTS-CGRU model outperforms the latest rumor events detection algorithms. (C) 2020 Published by Elsevier B.V.