《计算机应用研究》|Application Research of Computers

面向社交媒体的直接因果网络发现算法

Direct causal network discovery algorithm for social media

免费全文下载 (已被下载 次)  
获取PDF全文
作者 蔡瑞初,谢泳,陈薇,曾艳,郝志峰,杜文俊
机构 1.广东工业大学 计算机学院,广州 510006;2.佛山科学技术学院 数学与大数据学院,广东 佛山 528000;3.东北大学 工商管理学院,沈阳 110004
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2020)09-025-2689-05
DOI 10.19734/j.issn.1001-3695.2019.04.0159
摘要 高维时序因果网络发现是社交媒体因果关系发现的重要问题。然而,现有的时序因果关系发现方法不能发现直接因果以致因果网络推断结果不准确。针对此问题提出了一种直接因果网络发现方法。该方法考虑了时序因果模型的因果延迟、滞后期数量和条件节点集等因素,更准确地发现直接因果关系;另外,采用结合置换检验的因果关系检验方法,解决传递熵阈值难以设定的问题。实验结果表明,该方法在因果网络推断中优于现有方法,有效提升时序上直接因果网络推断的准确率,适用于发现潜在社交媒体因果关系网络。
关键词 因果关系; 时序; 社交媒体; 直接因果网络; 传递熵
基金项目 NSFC-广东联合基金资助项目(U1501254)
国家自然科学基金资助项目(61876043,61472089)
广东省自然科学基金资助项目(2014A030306004,2014A030308008)
广东省科技计划项目(2015B010108006,2015B010131015)
广东特支计划资助项目(2015TQ01X140)
广州市珠江科技新星资助项目(201610010101)
本文URL http://www.arocmag.com/article/01-2020-09-025.html
英文标题 Direct causal network discovery algorithm for social media
作者英文名 Cai Ruichu, Xie Yong, Chen Wei, Zeng Yan, Hao Zhifeng, Du Wenjun
机构英文名 1.School of Computers,Guangdong University of Technology,Guangzhou 510006,China;2.School of Mathematics & Big Data,Foshan University,Foshan Guangdong 528000,China;3.School of Business Administration,Northeastern University,Shenyang 110004,China
英文摘要 Time-series causal discovery for high-dimensional networks have been increasingly significant in social media causality. However, the existing algorithms' inability to discover direct causal relations renders the results of causal network infe-rence not so accurate. Hence, this paper proposed a direct causal network discovery algorithm. It considered various factors, including the causal delay, the lag length and the conditional nodes sets, in the time-series causal model to help improve the accuracy of direct causal network inferring. Further, the method solved the difficulty of setting transfer entropy thresholds through permutation test. Experimental results demonstrate that the method outperforms the existing algorithms in causal network inferring and can conspicuously improve the accuracy of direct causal network inference on time series, which is suitable for discovering potential causal networks in social media.
英文关键词 causality; time-series; social media; direct causal network; transfer entropy
参考文献 查看稿件参考文献
 
收稿日期 2019/4/25
修回日期 2019/6/18
页码 2689-2693
中图分类号 TP181
文献标志码 A