arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2312.02810cs.SIphysics.soc-ph
收藏

使用 SP!CE 框架对社交媒体影响活动进行编码:2022 年巴西总统选举案例研究

Using the SP!CE Framework to Code Influence Campaign Activity on Social Media: Case Study on the 2022 Brazilian Presidential Election

Alexander Gocso, Claudia Perez Brito, Bryan Ruesca, Allen Mendes, Mark A. Finlayson

AI总结:

本文以 2022 年巴西总统选举第二轮为案例,展示如何用 SP!CE 框架编码社交媒体影响活动,并发布帖子编码与截图。

摘要:

我们描述了一个案例研究,内容是使用 Structured Process for Information Campaign Enhancement(SP!CE,2.1 版)评估 2022 年 10 月巴西总统选举第二轮中存在的影响活动。SP!CE 是一个以美国军方为重点的框架,用于描述影响活动中的友方和对手行动,并且可与 Disinformation Analysis and Risk Management(DISARM)框架互操作。本案例研究的目的是展示如何使用 SP!CE 描述影响活动行为。我们选择巴西选举作为案例研究对象,因为众所周知,竞选期间社交媒体上存在大量错误信息与虚假信息。我们的目标是展示 SP!CE 如何应用于这种情境,说明社交媒体内容如何与信息活动行为对应,以及如何利用这种对应分析当时有哪些错误信息和虚假信息叙事在发挥作用。此外,我们旨在就如何在后续研究中应用该框架提供最佳实践方面的见解。我们公开了相关社交媒体帖子的编码和截图,以支持未来研究。

英文摘要:

We describe a case study in the use of the Structured Process for Information Campaign Enhancement (SP!CE, version 2.1) to evaluate influence campaigns present in the 2nd round of the Brazilian presidential election in 2022 October. SP!CE is a US-military focused framework for describing both friendly and adversary actions in influence campaigns, and is inter-operable with the Disinformation Analysis and Risk Management (DISARM) framework. The purpose of the case study is to demonstrate how SP!CE can be used to describe influence campaign behaviors. We selected the Brazilian election as the target of the case study as it is known that there were significant amounts of mis- and disinformation present on social media during the campaigns. Our goal was to demonstrate how SP!CE could be applied in such a context, showing how social media content could be aligned with information campaign behaviors and how such an alignment can be used to analyze which mis- and disinformation narratives were in play. Additionally, we aim to provide insights on best practices regarding how to apply the framework in further research. We release the coding and screenshots of the relevant social media posts to support future research.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑