Decoding the Narrative of Conspiracy Theories Using Machine Learning Model
A machine learning model distinguishes the conspiracy theory from the actual events and aids in understanding the narrative behind such theories.
For a while, the spread of conspiracy theories has become the new normal in social media. As the political and societal preferences are altering, the extent of such theories is getting accelerated. Before the USA elections this year, conspiracy theory QAnon dominated social media. The theory not only proved divisive but also led to the evolution of hate crimes amidst the historic presidential elections. More than one lakh accounts supporting QAnon were removed by Facebook and Twitter last month. Conspiracy theories like QAnon add burden to the pre-existing flaws ingrained in the society. With its divisive attributes, conspiracy theories seeds hate-mongering, communalism, and racism. According to the Merriam Webster dictionary, a conspiracy theory explains an event or set of circumstances as the result of a secret plot by usually powerful conspirators. Additionally, it targets only one cohort group that has the potential to alter the outcomes of an event.
The humongous challenge that regulatory authorities often face while dealing with conspiracy theories is navigating these theories’ origin. Certainly, these theories are based on a specific narrative that a particular group follows. But tracing such groups in social media becomes perverse as either the groups get dissipated, or the activities in such groups are fleeting. Additionally, as these groups communicated in encrypted languages, it becomes extremely difficult to understand the context of such communication. While regulatory authorities and social media executives have taken cognizance to curb this new mechanism that segments the society, researchers are working to trace the origin and narrative of such theories.
To acknowledge the challenges of conspiracy theories, researchers at the University of California have created an AI model that distinguishes between actual conspiracy theory from the real incidents. The research paper titled “An Automated Pipeline for the discovery of conspiracy and conspiracy theory narrative frameworks: Bridgegate, Pizzagate, and Storytelling on the web” cites that a graphical generative machine learning model can identify the context of posts on social media. The model has nodes that represent actors/actants, and multi-edges and self-loops among nodes capture context-specific relationships. The researchers have based their work on two separate comprehensive repositories of blog posts and news articles describing the well-known conspiracy theory Pizzagate from 2016, and the New Jersey political conspiracy Bridgegate from 2013.
Additionally, the researchers have captured context-specific actants and interactant relationships by developing a system of supernodes and subnodes. They have constructed an actant-relationship network, which constitutes the underlying generative narrative framework for each theory. Further, this model identifies the Pizzagate framework that relies on the conspiracy theorists’ interpretation of “hidden knowledge” to link otherwise unlinked domains of human interaction, which is an important feature of conspiracy theories. The model also contrasts this from the single domain focus of an actual conspiracy theory. It implies that Pizzagate spreads nationally, whereas Bridgegate was confined to only New Jersey politics.
Researchers have also highlighted the structural differences between the two narrative frameworks that private and public analysts can use to help distinguish between conspiracy theories and conspiracies.
What are Pizzagate and Bridgegate?
Evolved in the year 2016, Pizzagate is a false conspiracy theory that claimed Hilary Clinton and her former campaign chair, John Podesta, ran a pedophilia ring at the basement of a pizzeria in Washington DC. Followed by which, an armed man started shooting near the pizzeria at the Comet Ping Pong area of Washington DC. Fortunately, no casualties or human damage were reported. However, investigations revealed that Pizzagate was initiated on the troll haven and message board 4chan. Donald Trump supporters are accused of heavily circulating this conspiracy theory on social media before the 2016 USA elections.
Bridgegate is a conspiracy theory which revealed the discrepancies in the elections campaign during 2013 the New Jersey’s re-election. In September 2013, the port officials closed the lane between New York and New Jersey for several hours, citing that the governor wants to conduct a study about traffic in that particular lane. Investigations revealed that then New Jersey Governor Chris Christie had intentionally disrupted the traffic flow, as an act of political vengeance, infringing the constitutional provision.