社交网络假消息辟谣作用机理

doi:10.3785/j.issn.1008-973X.2021.04.003

浙江大学学报(工学版)

2021, Vol. 55

Issue (4): 615-625 DOI: 10.3785/j.issn.1008-973X.2021.04.003

计算机技术、电信技术

社交网络假消息辟谣作用机理

张玉琪(

),郭斌*(

),丁亚三,刘思聪,於志文

西北工业大学计算机学院，陕西西安 710129

Mechanism of corrections to false information in online social network

Yu-qi ZHANG(

),Bin GUO*(

),Ya-san DING,Si-cong LIU,Zhi-wen YU

School of Computer Science, Northwestern Polytechnical University, Xi’an 710129, China

全文: PDF(1191 KB) HTML

摘要：

研究真实社交网络环境下假消息辟谣作用机理. 提出评估辟谣效果的方法及探究影响辟谣效果的因素. 基于已有研究成果与假设，总结出8个影响辟谣效果的因素，如原假消息内容占比、是否包含谣言文字警示、是否解释原因、用户影响力等. 使用情感分析和微博社交上下文，评估辟谣微博的辟谣效果. 利用统计学方法，检验预设影响因素与辟谣效果间的关系. 基于新冠疫情相关的辟谣微博数据开展实验，实验分析表明，辟谣信息中原假消息内容占比和辟谣效果呈负相关，解释原因与辟谣效果呈正相关. 提出尽量少地提及原假消息、应解释原假消息错误的原因等6条辟谣建议，为社交网络假消息辟谣提供指导.

关键词： 假消息; 辟谣机理; 情感分析; 社交网络; 微博

Abstract:

The mechanism of corrections on false information in real social network environments was analyzed. The effect of corrections was evaluated and its influencing factors were explored. Eight factors that affect the effectiveness of correction were summarized based on existing research and our hypotheses, such as the proportion of the original false information, whether it contains text warnings of false information, whether to explain the explanation, user influence, etc.. The effectiveness of correction posts was evaluated by sentiment analysis and the social context of themselves. Statistical methods were used to test the relationship between the pre-determined influencing factors and the effectiveness of correction. The experiment was conducted based on the false information data about COVID-19 epidemic collected from Sina Weibo. Results show that a higher proportion of false information in a correction reduces the effectiveness, and explaining the reason improves the effectiveness. Six conclusions that improve the effectiveness of corrections on social networks were proposed such as mentioning original misinformation less, explaining why original misinformation is wrong. Guidance was provided for related media to correct false information on social network.

Key words: false information mechanism of correction sentiment analysis social network Weibo

收稿日期: 2021-01-27 出版日期: 2021-05-07

CLC:

TP 399

基金资助: 国家重点研发计划资助项目（2019QY0600）；国家自然科学基金资助项目（61772428，61725205）

通讯作者: 郭斌 E-mail: 1347088657@qq.com;guobin.keio@gmail.com

作者简介: 张玉琪（1998—），女，博士生，从事数据挖掘的研究. orcid.org/0000-0001-8173-3844. E-mail： 1347088657@qq.com

	服务
	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	作者相关文章
	张玉琪
	郭斌
	丁亚三
	刘思聪
	於志文

引用本文:

张玉琪,郭斌,丁亚三,刘思聪,於志文. 社交网络假消息辟谣作用机理[J]. 浙江大学学报(工学版), 2021, 55(4): 615-625.

Yu-qi ZHANG,Bin GUO,Ya-san DING,Si-cong LIU,Zhi-wen YU. Mechanism of corrections to false information in online social network. Journal of ZheJiang University (Engineering Science), 2021, 55(4): 615-625.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2021.04.003 或 http://www.zjujournals.com/eng/CN/Y2021/V55/I4/615

表 1 常见假消息分类表

图 1 辟谣微博示例

图 2 辟谣机理探究框架图

表 2 微博数据格式

表 3 评论数据格式

表 4 影响辟谣因素归纳表

表 5 人工标注说明

表 6 语料数据格式

表 7 数据预处理过程

图 3 模型指标的对比

表 8 不相关评论示例

图 4 情感分类流程图

表 9 预设影响因素的数据类别

表 10 关系分析方法选取

表 11 差异分析结果

表 12 斯皮尔曼相关分析结果

表 13 谣言文字警示、图片警示以及包含图片或文字警示的频率分布

图 5 认证用户的分布

图 6 是否解释原因的辟谣效果对数分布的对比

图 7 原假消息内容占比、信息来源影响力与辟谣效果的关系

图 8 原假消息内容占比与来源影响力对辟谣效果共同作用

图 9 辟谣效果、辟谣数目以及来源影响力的时序变化

1	SHU K, BERNARD H R, LIU H. Studying fake news via network analysis: detection and mitigation[M]// Emerging Research Challenges and Opportunities in Computational Social Network Analysis and Mining. Cham: Springer, 2019: 43–65.
2	LAZER D M J, BAUM M A, BENKLER Y, et al The science of fake news[J]. Science, 2018, 359 (6380): 1094- 1096 doi: 10.1126/science.aao2998
3	NGUYEN N P, YAN G, THAI M T, et al. Containment of misinformation spread in online social networks [C]// Proceedings of the 4th Annual ACM Web Science Conference. Wuhan: ACM, 2012: 213-222.
4	TRIPATHY R M, BAGCHI A, MEHTA S, et al. A study of rumor control strategies on social networks [C]// Conference on Information and Knowledge Management. Toronto: ACM, 2010: 1817-1820.
5	JERIT J Partisan perceptual bias and the information environment[J]. The Journal of Politics, 2012, 74 (3): 672- 684 doi: 10.1017/S0022381612000187
6	BUSH J G, JOHNSON H M, SEIFERT C M. The implications of corrections: then why did you mention it [C]// Proceedings of the 16th Annual Conference of the Cognitive Science Society. Atlanta: Elsevier, 1994: 112-117.
7	COOK J, LEWANDOWSKY S. The debunking handbook [EB/OL]. (2012-01-23). http: //sks.to/debunk.
8	LEWANDOWSKY S, ECKER U K, SEIFERT C M, et al Misinformation and its correction continued influence and successful debiasing[J]. Psychological Science in the Public Interest, 2012, 13 (3): 106- 131 doi: 10.1177/1529100612451018
9	SHARMA K, QIAN F, JIANG H, et al Combating fake news: a survey on identification and mitigation techniques[J]. ACM Transactions on Intelligent Systems and Technology, 2019, 10 (3): 1- 42
10	BUDAK C, AGRAWAL D, ABBADI A E, et al. Limiting the spread of misinformation in social networks [C]// Proceedings of the 20th International Conference on World Wide Web. Bangalore: ACM, 2011: 665-674.
11	HE X, SONG G, CHEN W, et al. Influence blocking maximization in social networks under the competitive linear threshold model [C]// Proceedings of the 2012 SIAM International Conference on Data Mining. Anaheim: SIAM, 2012: 463-474.
12	FARAJTABAR M, YANG J, YE X, et al. Fake news mitigation via point process based intervention [C]// International Conference on Machine Learning. Sydney: ML Research Press, 2017: 1097-1106.
13	AMORUSO M, ANELLO D, AULETTA V, et al Contrasting the spread of misinformation in online social networks[J]. Journal of Artificial Intelligence Research, 2020, 69: 847- 879 doi: 10.1613/jair.1.11509
14	YU S, VOROBEYCHIK Y, ALFELD S, et al. Adversarial classification on social networks [C]// Adaptive Agents and Multi-agents Systems. Stockholm: ACM, 2018: 211-219.
15	THORSON E A Belief echoes: the persistent effects of corrected misinformation[J]. Political Communication, 2016, 33 (3): 460- 480 doi: 10.1080/10584609.2015.1102187
16	NYHAN B, REIFLER J When corrections fail: the persistence of political misperceptions[J]. Political Behavior, 2010, 32 (2): 303- 330 doi: 10.1007/s11109-010-9112-2
17	KIM A, DENNIS A R. Says who?: how news presentation format influences perceived believability and the engagement level of social media users [C]// Proceedings of the 51st Hawaii International Conference on System Sciences. Hawaii: AIS eLibrary, 2018: 1-11.
18	BODE L, VRAGA E K In related news, that was wrong: the correction of misinformation through related stories functionality in social media[J]. Journal of Communication, 2015, 65 (4): 619- 638 doi: 10.1111/jcom.12166
19	BODE L, VRAGA E K See something, say something: correction of global health misinformation on social media[J]. Health Communication, 2018, 33 (9): 1131- 1140 doi: 10.1080/10410236.2017.1331312
20	PIERRI F, CERI S False news on social media: a data-driven survey[J]. Sigmod Record, 2019, 48 (2): 18- 27 doi: 10.1145/3377330.3377334
21	SCHEUFELE D A, KRAUSE N M Science audiences, misinformation, and fake news[J]. Proceedings of the National Academy of Sciences of the United States of America, 2019, 116 (16): 7662- 7669 doi: 10.1073/pnas.1805871115
22	LEWANDOWSKY S, ECKER U K, COOK J, et al Beyond misinformation: understanding and coping with the “post-truth” era[J]. Journal of Applied Research in Memory and Cognition, 2017, 6 (4): 353- 369 doi: 10.1016/j.jarmac.2017.07.008
23	SKURNIK I, YOON C, PARK D C, et al How warnings about false claims become recommendations[J]. Journal of Consumer Research, 2005, 31 (4): 713- 724 doi: 10.1086/426605
24	SCHWARZ N, SANNA L J, SKURNIK I, et al Metacognitive experiences and the intricacies of setting people straight: implications for debiasing and public information campaigns[J]. Advances in Experimental Social Psychology, 2007, 39 (6): 127- 161
25	NYHAN B, REIFLER J The roles of information deficits and identity threat in the prevalence of misperceptions[J]. Journal of Elections, Public opinion and parties, 2019, 29 (2): 222- 244 doi: 10.1080/17457289.2018.1465061

[1]	王亮, 於志文, 郭斌, 熊菲. 基于移动社交网络的群智感知社群化任务分发[J]. 浙江大学学报(工学版), 2018, 52(9): 1709-1716.
[2]	孙念, 李玉强, 刘爱华, 刘春, 黎威威. 基于松散条件下协同学习的中文微博情感分析[J]. 浙江大学学报(工学版), 2018, 52(8): 1452-1460.
[3]	张亚楠, 陈德运, 王莹洁, 刘宇鹏. 基于增量图形模式匹配的动态冷启动推荐方法[J]. 浙江大学学报(工学版), 2017, 51(2): 408-415.
[4]	欧阳逸, 郭斌, 何萌, 於志文, 周兴社. 微博事件感知与脉络呈现系统[J]. 浙江大学学报(工学版), 2016, 50(6): 1176-1182.
[5]	张亚楠, 曲明成,刘宇鹏. 基于社交关系拓扑结构的冷启动推荐方法[J]. 浙江大学学报(工学版), 2016, 50(5): 1001-1008.
[6]	林萌, 罗森林, 贾丛飞, 韩磊, 原玉娇, 潘丽敏. 融合句义结构模型的微博话题摘要算法[J]. 浙江大学学报(工学版), 2015, 49(12): 2316-2325.
[7]	刘芳, 孙芸, 杨庚, 林海. 基于粒子群优化算法的社交网络可视化[J]. J4, 2013, 47(1): 37-43.

Viewed

Full text

Abstract

Cited

Shared

Discussed