Research Associate Professor of Computer Science and Principal Scientist at USC Information Sciences Institute
- Doctoral Degree, Physics, University of California - Santa Barbara
- Bachelor's Degree, Physics, Princeton University
Kristina Lerman is a Project Leader at the Information Sciences Institute, a unit of the University of Southern California (USC), and a Research Associate Professor in the USC Viterbi School of Engineering's Computer Science Department. An expert in complex multi-agent systems, Dr. Lerman has received numerous grants on social data and other topics from the National Science Foundation (NSF), Defense Advanced Research Projects Agency (DARPA) and Information Sciences Institute (ISI) early in her career.
Her current work revolves around deciphering the structure and dynamics of "social web" sites such as Twitter, Digg, Flickr and Delicious. Among her goals: automatically organize collective knowledge, discover the structure of user-generated communities, and predict emerging trends and group behavior. Her empirical and experimental studies identified the importance of cognitive biases to understanding individual and collective behavior online.
Dr. Lerman is an active member of the social computing research community, as a chair (SocInfo’14, SocialCom’14, Hypertext’13) or senior PC (ICWSM, IJCAI) of leading conferences, and has organized several computational social science workshops.
Dr. Lerman teaches a USC Computer Science Department course on Social Media Analysis.
Much of my current research focuses on the social Web, a label that includes social media and online social networking sites. These technologies have transformed the Web into an active medium, where people create, annotate, evaluate and distribute information. The social Web is an exciting, complex domain for developing new AI algorithms and applications that leverage people's interactions as a platform for computing. My group studies novel behaviors and structures that emerge when people collectively organize and use information.
Learning from Social Metadata
Many social Web sites allow users to create content and annotate it with descriptive metadata, such as tags, and to organize content within personal hierarchies. Structured social metadata offers invaluable evidence for learning how a community organizes knowledge. But such metadata also tends to be sparse, shallow, ambiguous, noisy and inconsistent. We are developing machine learning methods to aggregate social metadata to improve information discovery and learn common taxonomies (folksonomies.)
Social Dynamics and Networks
As people interact on the social Web, their activity affects the structure of the Web itself, with complex feedback between individual and collective decisions producing qualitatively new online behaviors. We are developing a mathematical framework to model collective behavior, and to use these models for predicting trends, such as news story popularity. We are also developing novel metrics for network analysis and studying how network structure affects information spread.
Semantic Modeling of Information Sources
My research as a member of ISI's Information integration group deals with automatically recognizing semantics of data types used by various information sources. We use machine learning methods to represent and learn the structure of data extracted from information sources, and apply the learned representations to recognize new instances of the same data.
- 2003 USC Information Sciences Institute ISD Research Award
- 1999 USC Information Sciences Institute ISD Research Award
- 1991 UC Santa Barbara Patricia Roberts Harris Fellowship,
- 1989 UC Santa Barbara Dupont Nemurs Award,
- 1986 United Federation of Teachers Scholarship