Social media has been gradually emerging at the forefront of election campaigning, and this transition has been further accelerated with the ongoing pandemic. At TCPD, we are studying election campaigning on two major social media platforms – Facebook and Twitter. Facebook is uniquely situated in being the only major social media platform that permits paid political advertising, while Twitter is a commonly used platform for political discourse in India.
The objective of our study is to analyze political campaigning on social media at the level of contesting candidates. The study would be more granular than previous studies by looking at individual candidates, and thus being able to provide insights at all levels, i. e., trends within and across constituencies, within and across parties etc. Unlike typical studies in this domain, we employ a mixed-methods approach, using ML-assisted human annotations for creating the data needed for qualitative analysis, and machine learning models to further analyse this data. This is particularly a requirement for studying Facebook ads, because the content on Facebook is predominantly visual, making it difficult to run automated ML methods to classify the content.
We have also engaged with TTC Labs at Facebook during their Ads Targeting Transparency Design Jam workshop, which focused on better transparency on how political ads are designed and run on Facebook.
Key Researcher: Neelesh Agrawal