Meet Lynnette Ng — our Societal Computing PhD scholar at Carnegie Mellon University
We catch up with Lynnette, our DSTA PhD Scholar and Knight Scholar, currently pursuing her PhD in Societal Computing at Carnegie Mellon University. Passionate about computational social sciences and social media analysis, Lynnette develops advanced models to detect and analyze bot behavior. Her research also incorporates techniques like agent-based modeling to predict and mitigate the impact of these automated agents. Lynnette’s goal is to create effective strategies to combat disinformation and ensure the integrity of online information.
Hi, I’m Lynnette Ng, and I’m currently pursuing my PhD at Carnegie Mellon University. When I’m not buried in research on solving various challenges in social media bots and disinformation modelling, you can find me playing Violin for CMU’s University Orchestra or travelling on adventures around the US.
Could you share what inspired you to pursue this area of research?
My inspiration came from the fascinating interactions between social media bots and humans. These digital entities influence conversations and shape public opinion. By modeling bot behavior and its subtypes, we can detect these automated agents and counteract malicious ones.
This is crucial in the social media domain, where ethical experiments are challenging due to the dynamic nature of information flow. Modeling disinformation spread strategies allows us to predict and analyze the impact of harmful bots. These insights help analysts understand online social processes and inform policy decisions.
My research at CMU aims to identify the core structures of bot-driven disinformation, recognize patterns in real-world data, and study network effects through time-series simulations. This work enhances our ability to combat disinformation and maintain the integrity of online information.
What are some of the key findings from your research on social media bots?
A significant portion of my research centers on bot detection and behavior modeling. I developed advanced algorithms, such as BotBuster, which uses deep learning to identify and classify bots across multiple social media platforms. This tool has been effective in uncovering the dynamics of disinformation spread, particularly during critical events like the COVID-19 pandemic and political elections. My findings reveal that while influential humans often initiate disinformation, bots play a crucial role in amplifying these narratives, thereby distorting public discourse.
How does your research on bots relate to real-world events like elections?
In my recent publication, “Assembling a Multi-Platform Ensemble Social Bot Detector with Applications to US 2020 Elections”, I highlighted my approach to bot detection. By analyzing bot activities during the US 2020 elections, I demonstrated how coordinated bot networks can influence electoral outcomes, posing a significant threat to democratic processes.
Can you tell us more about your work on cyborgs and their role in social media?
Another fascinating aspect of my research is detailed in “Cyborgs for Strategic Communication on Social Media”. This study delves into the hybrid nature of cyborgs — accounts that exhibit both human and bot-like behaviors. These entities blend automated efficiency with a human touch to engage effectively with their audience, blurring the lines between human and automated actors.
What have you discovered about geopolitical disinformation strategies?
In my study, “Popping the Hood on Chinese Balloons: Examining the Discourse Between U.S. and China-Geotagged Accounts”, I provided insights into geopolitical disinformation strategies. By examining interactions between US and China-based accounts, I uncovered the nuanced ways in which bots and human actors collaborate to propagate certain narratives. This study is particularly relevant in the context of international relations, where information warfare plays a pivotal role.
How does your research address the offline impact of online propaganda?
My interdisciplinary research on “Online to Offline Crossover of White Supremacist Propaganda” explores the transition of extremist content from digital platforms to real-world actions. Conducted in collaboration with other researchers, this study highlights the dangerous implications of coordinated online propaganda and its potential to incite offline violence.
What do you enjoy doing outside of your research?
When I’m not immersed in my papers, I love playing the violin — I’m part of CMU’s Orchestra. Additionally, I enjoy exploring Pittsburgh and the wider United States as an avid nature photographer. The freedom to explore and experience new places is something I truly cherish, especially after overcoming significant health challenges.
Can you share what’s on your reading list right now?
Person, Thing, Robot: A Moral and Legal Ontology for the 21st Century and Beyond by David J. Gunkel.
Could you point us to recent papers that you’re especially proud about?
[1] Feb 2024 Presented Cyborgs for strategic communication on social media published at Big Data and Society! Read it here
[2] Jan 2024 Presented Challenges for Real-Time Toxicity Detection in Online Games at ACM Ethical Games Conference Watch the talk here here
[3] Nov 2023 Paper It Takes Two to Negotiate: Modeling Social Exchange in Online Multiplayer Games is accepted at CSCW 2024! Read the preprint here here
[4] Sept 2023 Paper Do you hear the people sing? Comparison of synchronized URL and narrative themes in 2020 and 2023 French protests is published in Frontiers of Big Data. I wrote this paper in the cardiac ICU because I thought my life was at an end. Read it here
[5] Jan 2023 Paper A combined synchronization index for evaluating collective action social media is published in Applied Network Science Read it here
Check out the rest of Lynnette’s work on her Research Portfolio, and daily glimpses of her PhD life and trips on Instagram.
