Carnegie Mellon University
School of Computer Science  ›  Software and Societal Systems Department  ›  People  ›  core-faculty  ›  Kathleen Carley

Kathleen Carley

Dr. Kathleen M. Carley

Professor of Computer Science, Director of the Center for Informed Democracy and Social-cybersecurity (IDeaS) and Director of the Center for Computational Analysis of Social and Organizational Systems (CASOS), Software and Societal Systems

Address
5000 Forbes Avenue
Pittsburgh, PA 15213

Bio

Kathleen M. Carley is a professor in the Software and Societal Systems Department in Carnegie Mellon's School of Computer Science. She also has courtesy appointments at:

Dr. Carley’s research combines cognitive science, sociology, organization science, and computer science to address complex social and organizational issues such as disinformation, counter-terrorism, and organizational design. In her research she typically employs network science, machine learning, computational linguistics, and agent based modeling methods. Her most notable research contributions are the establishment of Dynamic Network Analysis (DNA) and the associated theory and methodology for examining large high‐dimensional time variant networks, and Social-Cybersecurity (SC) and the associated theory and methodology for identifying and mitigating online harms and adverse actors. Her research on DNA and SC has resulted in tools for analyzing large‐scale dynamic network, agent-based computer simulation systems, and tools for bot-detection, hate-speech detection, and scenario creation for live virtual constructive exercises. She has led the development of tools for extracting sentiment, social and semantic networks, and cues from textual data (AutoMap & NetMapper), simulating epidemics (BioWar), influence operations and information diffusion (Construct), assessing networks (including this in social media (ORA), creating scenarios (AESOP), and validating synthetic social media data (MOMUS). Her ORA system is one of the premier network analysis and visualization technologies. It is used worldwide and supports reasoning about geo‐spatial and dynamic high‐dimensional network data. It includes special features for social media data, citation data, supply chains and network dynamics. Illustrative projects include assessment of disinformation and social cyber‐security threats, IRS outreach, airline re‐rerouting, counter-terrorism, counter‐narcotics, insider threat, health analytics, social media analytics of elections around the globe, and crises such as Benghazi, Darfur, the Arab Spring, COVID‐19.

Education

I have and SB degrees in economics and a second SB in political science from M.I.T., a Ph.D. degree in sociology from Harvard University, and an HD from the University of Zurich. While at MIT and Harvard I took every PhD level AI course offered. I teach dynamic social network analysis (17-685/17-801)each spring, which takes an interdisciplinary approach to network science. I also teach a course on simulation(17-621/17-821) every other fall, where we cover agent bases modeling and system dynamics. Both are project courses. Master's students and advanced undergraduates should sign up for 17-685 or 17-621 and Ph.D. students for 17-801 or 17-821. For an independent study on social media analytics or disinformation use 17-996.

Research

Research Areas of Interest

  • Complex Socio-Technical Systems
  • Computing Technology and Policy
  • Network Science and Social Networks
  • Organizations
  • Software Data Analysis

Projects

Project OMEN
Project Omen is a large multi-institution project for which I am the chief scientist. The goal of project OMEN is to create and deliver new technology for understanding influence and information operations in the online environment, a persistent training and planning range where this and other technology can be trained on and used, and the synthetic data to support that training and planning. This training covers online social-cyber safetey activities such as - identification and responding to disinformation, recognizing inauthentic behavior (bots) and responding to it, identifying influence activities, and creating influential messaging. Tools within this include: AESOP - a scenario generator, SynTel - a synthetic LLM based telegram generator, MOMUS - an evaluation system for synthetic data, Artemis - a network science+AI tool for assessing the quality of LLM generated synthetic data.

BEND
Assessing the Information Environment: In this project the goal is to improve our ability to identify who is trying to influence whom in social media and across social media platforms and to what effect. It is centered around 8 maneuvers for influencing what is being said and how, and 9 for influencing who is saying what, what groups exist, and who leads them. This effort seeks to employ an improved understanding of how the social position, the identify, the emotional state and the moral values of the send influence what messages they send and with what motive and in identifying how the content of those messages impacts those who read them. One aspects of this project is to generate an improved theory and methods for assessing cognitive-emotional agency and the way people change their identities in response to influence and in-order to influence others. Another aspect of this project is to apply the new methods and theory to assess adversarial and US activity in the Asia-Pacific region, the Ukraine, and during large scale disasters such as COVID.

Scalable Social-Cybersecurity Technologies 
This project focuses on developing new bot, troll, hate-speech, disinformation, and other social0cyber tools that are scalable, do not require GPUs, work cross-culturally and in multiple languages, work on data from multiple social media platforms, and are based on foundational social, cultural and cognitive insights. Example tools are bot-hunter and the hate-speech detector.

Cyber-security & Human Behavior 
This progress seeks to understand the interaction between human behavior and cyberattacks. For this project - three agent-based simulations are being developed - OSIRIS -- a digital twin of an organization (that can be the target of a cyber attack), CYBER-FIT - a model of the cyber team within the organization that responds to cyber-attacks and tries to prevent, stop or mitigate the consequences of such attacks, and RED-CYBER a lite system that generates attacks. In OSIRIS and CYBER-FIT the agents are subject to human related issues such as education, cyber-awareness, fatigue, stress, and fear. This influences their actions and so their susceptibility to attacks - which in turn influences the success of the attack. Numerous virtual experiments are being run to understand the impact of human factors, and to identify successful interventions.

Publications

Lynnette Hui Xian Ng, Ian Kloo, Samantha Clark, Kathleen M. Carley, 2024, “An exploratory analysis of COVID bot vs human disinformation dissemination stemming from the Disinformation Dozen on Telegram”. Journal of Computational Social Science, 7, 695–720, DOI: https://doi.org/10.1007/s42001-024-00253-y

Daniele Bellutta, Kathleen M. Carley, 2024, Indicators of the formation of precedent at the International Court of Justice, Social Networks, vol 79:1-13. DOI: https://doi.org/10.1016/j.socnet.2024.04.001

Christine Sowa Lepird, Lynnette Hui Xian Ng, Kathleen M. Carley, 2024, "Non-credibility scores: Relative ranking of news sites shared on social media to identify new pink slime sites",First Monday, Sept 2024.

Charity Jacobs, Lynnette Hiu Xian Ng, Kathleen M. Carley, 2024, “DET: Detection Evasion Techniques of State-Sponsored Accounts” Workshop Proceedings of the 18th International AAAI Conference on Web and Social Media, DOI: 10.36190/2024.63

Samantha C. Phillips, Lynnette Hui Xian Ng, Wenqi Zhou, and Kathleen M. Carley, 2024, Moral and emotional influences on attitude stability towards COVID-19 vaccines on social media, In proceedings of the International Conference on Social Computing, Behavioral-Cultural Modeling & Prediction and Behavior Representation in Modeling and Simulation Conference, SBP-BRiMS 2024, Eds Robert Thomson, Aravind Hariharan, Scott Renshaw, Samer Al-khateeb, Annetta Burger, Patrick Park & Aryn Pike (Eds) September, 2024, Pittsburgh PA. DOI: https://doi.org/10.48550/arXiv.2407.19406 Best Paper Runner Up

Joshua Uyheng, Dawn C. Robinson, and Kathleen M. Carley, 2023, “Bridging online and offline dynamics of the face mask infodemic,” BMC Digital Health 1, 27 (2023). https://doi.org/10.1186/s44247-023-00026-z

Jeongkeun Shin, L. Richard Carley, and Kathleen M. Carley, 2024-forthcoming “Simulation-Based Study on False Alarms in Intrusion Detection Systems for Organizations Facing Dual Phishing and DoS Attacks.” In 2024 Annual Modeling and Simulation Conference (ANNSIM), Best Paper Runner Up

Jeongkeun Shin, L. Richard Carley, Geoffrey B. Dobson, and Kathleen M. Carley, 2023. “Modeling and Simulation of the Human Firewall Against Phishing Attacks in Small and Medium-Sized Businesses.” In 2023 Annual Modeling and Simulation Conference (ANNSIM), pp. 369-380. IEEE, 2023.

David M. Beskow and Kathleen M. Carley, 2019, Social Cybersecurity: An Emerging National Security Requirement, Military Review, March-April2019 – see https://www.armyupress.army.mil/Journals/Military-Review/English-Edition-Archives/Mar-Apr-2019/117-Cybersecurity/b/.