Improving the health of our public sphere through conversations
Source: https://www.cortico.ai
It go without saying that we currently live in an extremely divisive and hostile political climate. Even little differences can magnify the intensity of discourse between two people with opposing views. When we think about social networks, we can begin to rationalize reasons that may explain the magnitude of these frustrating sentiments.
The above image depicts the polarization between people on social media and the underlying conclusion is that people overwhelmingly interact only with those who share the same perspectives as them, with very little crossover between the two groups. As such, people are necessarily unable to understand others’ sides because they rarely get to even listen to it.
When Trump won the presidential election in 2016, seeking to understand the how and the why of how he got into the White House became an a huge topic of interest for a lot of people. Suddenly, it became important to break across these social network barriers to really gauge the perspectives and rationales of those who got him there. It was also at this time that my mentor, Eugene Yi, began to work on what would become his start up company, Cortico.
At the heart of Cortico is the mission to provide journalists with access to genuine and meaningful stories of people in the United States. The approach to doing so involves a mix of localized search of public conversation and community engagement. The former is done through Cortico’s Earshot, which uses machine learning algorithms to capture and organize conversations between people whether it be through television, radio, or social media. The latter provides a means for people to simply share their own stories and opinions.
Whether it be the problem that Cortico tries to address or the solutions through which it does so, Cortico finds relevance in a lot of what we discuss in class. The topic of social networks and concentration of interactions within certain clusters is directly demonstrated in the segregation of social media interaction between people. At the same time, in order for Cortico to actually break down this segregation, it needs to capture the discussions regarding key issues by regular people. To do so, the use of networks is once again applicable as it would need to track these conversations within social networks both within clusters and across them to gauge all sides.