Srijan Kumar, Assistant Professor,
School of CSE
Can you describe how your recent research is helping safeguard Asian communities online
Our research began in mid-March of 2020 which is essentially when the COVID-19 lockdown first began in the United States. At this time, we started to see media reports of harassment targeting Asians and Asian Americans, not just in the U.S. but also around the world. To understand this phenomenon and combat it, we undertook the largest study of anti-Asian rhetoric in online social media. We created three classifiers to determine the prevalence of anti-racism and hate targeting people of Asian descent on Twitter. It is important to assess online harassment because social media and web platforms have become such an integral part of our life. Most of our information, discussions, political opinions and beliefs are formed with who we interact with online and what we read online. Online chatter provides a lens to characterize large-scale opinions in real-time which influences incidents happening in the real world (e.g., the Capital riots in January 2021 were orchestrated via social media platforms). If we can understand and prevent harassment digitally, we can potentially prevent it physically.
What inspired you to begin researching human behavior on social media?
The reason I got into this particular area was because I grew up with social media and I was quite fascinated in understanding human behavior at scale. I wanted to understand the good, bad, and the ugly of the Internet which became a big focus of what my research over the years including what my lab, CLAWS, works on.
Speaking of your lab, are you currently accepting any more researchers on your team?
We are always looking for highly motivated students who want to do research that impacts the development of society! We use large-scale data and computational methods to revolutionize the next generation of the internet, web, and social media.
Students who are interested in working in the CLAWS lab can apply here.
What are some recent research examples that help society by analyzing online communities?
One project I would like to highlight is on health misinformation, which is essentially prevalent on social media and a menace to society at large. Here, with my Ph.D. student Bing He, we have been looking at how misinformation can be countered and prevent it from spreading. So, if we enable people to speak out against misinformation when they see it, that can have a huge impact.
Another research project I would like to highlight is led by CSE Ph.D. student Sejoon Oh who is working on recommendation systems. A lot of what we see and read online is essentially powered by recommendation systems. And a lot of what we do in the real world, such as choosing which restaurant or doctor to go to, is in some way or another influenced by these recommendation systems. So, we are looking at the accuracy and trustworthiness of recommender systems so that we can create the next generation of recommendation systems so that the best information can be provided to people.
How has your approach to looking at social media users changed over the years?
We are now looking at the entire information system at large and the actors that are part of it. However, I did first begin exploring this question by looking at the bad actors on the internet. As a Ph.D. I began trying to understand vandalism on Wikipedia, trolling behavior on social networks, people using multiple accounts on social platforms, and people posting misinformation on Wikipedia. Since then, we have expanded and have begun looking at the good side as well such as looking at how to create better recommendation systems, how to counter misinformation and see what we can do to make the online platforms better. These are very important problems and the web and social media are ever-evolving entities so there will always be problems that need to be solved. This means there is a need for researchers, practitioners, and experts to work on these problems so that together, we can find actionable solutions.