The CSIR-DSFSI-MMZA 2021 Municipal Election Social Media Monitoring Project

Contact

You can contact Dr. Vukosi Marivate for enquiries, comments or questions: vukosi.marivate@cs.up.ac.za

Introduction to the Project

The growing use of smartphones and social media platforms such as Twitter has exacerbated the spread of misinformation and disinformation significantly. People use these platforms to keep in touch, share information and keep up to date with what is happening locally and around the world. With the various sources of information it's becoming more and more challenging for people to distinguish between what is true and what is false[^1]. This was more apparent when the first COVID-19 case was reported. People started sharing information on these platforms, their concerns and different opinions about COVID-19 and seeking information about the pandemic. With that, a lot of mis- or dis- information was shared and it became increasingly difficult for people to distinguish which of the information was true. Thus, leading to issues like vaccine hesitancy etc. The issue of misinformation and disinformation is not only prevalent in cases of the pandemic. However as notable events took place around the world, particularly elections, the threat of misinformation and disinformation grew due to online platforms.

This caused the false content to spread more rapidly and reach a wider audience during election events despite the efforts of different social media companies trying to curb it; by banning rogue accounts, removing false content, etc[^2]. Many online users are still able to share different forms of this content such as deep fakes, conspiracy theories, misleading stories, etc., such that if not verified or reviewed by the consumer it can lead to confusion, doubt, panic and anger resulting to violence, voters being intimidated or influenced, etc. It was witnessed during the 2020 United States presidential elections that social media platforms were used to share manipulated content, disinformation campaigns and bot networks were used to disrupt the election process, and to incite violence[^3].

Other countries such as Brazil, Ghana experienced similar challenges. In South Africa the local government elections will take place on 1 November 2021, and as this day approaches we're all preparing for this time in various ways, e.g. finding ways to tackle the spread of mis- and disinformation. In this project we conduct research to monitor and investigate if instances of mis- or disinformation can be detected early from a Twitter dataset consisting of discourses about the elections, these discourses are between the general populace and political leaders.

The Team

CSIR Information and Cyber security Center

The CSIR Information and Cyber Security Center (ICSC) under the Defence and Security cluster, focuses on research, development, and innovation of home-grown cyber and information security; inclusive of identity management solutions and approaches to securely identify and protect people and technological systems (physical and digital) against vulnerabilities, threats and risks in the digital realm. This is achieved through partnerships and collaborations with public and private sectors.

CSIR ICSC Team Members

  • Nelisiwe Dlamini
  • Sthembile Mthethwa
  • Lungisani Ndlovu
  • Ntombizodwa Thwala

Data Science for Social Impact (DSFSI@UP)

Data Science for Social Impact Research Group at the Computer Science Department, University of Pretoria. Our general areas of work straddle Data Science for Society as well as Local Language Natural Language Processing. These two strands are complementary. Our work in Data Science and Society has allowed us to have a more nuanced approach to understanding the systematic challenges that face being able to do excellent science with local languages. Through Data Science for Society, we have to understand how when one carries through Data Science research, we situate how the users are part of the process. We find that we need to adjust our research to take care of these challenges and innovate in ways we gather direct data or alternative data.

DSFSI Team Members

  • Vukosi Marivate
  • Herkulaas MVE Combrink
  • Mohlatlego Nakeng
  • Temitope Kekere
  • Thapelo Sindane
  • Tshepiso Mokoena

MEDIA MONITORING AFRICA (MMA)

Media Monitoring Africa (MMA) is a Non-governmental organization (NGO) operating since 1993. MMA has monitored media coverage of all South Africans democratic elections. In 2019 MMA worked with the Electoral Commission (IEC) on a world first public complaints system to help combat and mitigate mis and disinformation. The Real411 platform has since evolved and with nearly 2000 complaints it is once again with the IEC the official platform for the public to submit complaints about mis and disinformation in the lead up to the local government elections (LGE) 2021. Each complaint is analysed by a minimum of 4 experts which means that the examples of mis and disinformation that can be used for this exciting collaboration have great value as curated data.

MMA Team members

  • William Bird (MMA Director)
  • Thandi Smith

Supported by our Developer team at Assemble.

A study of election information through Twitter - Our Analysis

Website Assets

  • Graph by Nicholas Menghini licensed under CC BY 3.0 is used as a favicon.