IC Colloquium - Reshaping Terrorist Networks
Event details
| Date | 11.11.2013 |
| Hour | 16:15 › 17:30 |
| Speaker | VS Subrahmanian - University of Maryland |
| Location | |
| Category | Conferences - Seminars |
Abstract:
Though terrorist networks have been extensively studied, there has been limited work on destabilizing them. We look at the problem of removing "k" people from a terror network (e.g. by capturing them, buying them off, relocating them) in order to minimize the operational effectiveness of the network. We solve this via 3 steps: (i) we develop a framework to predict who succeeds a terrorist when he is ``removed''. (ii) When a set of terrorists is removed, we show how to infer a probability distribution on the resulting possible new networks. (iii) Using the solutions to the previous two problems, we find a set of "k" people to remove from a terror network so that the expected effectiveness of the resulting network is minimized. We test our solutions to the first problem using (i) a synthetic data set and (ii) data sets on terror networks associated with 4 terror groups. For the third problem, we test our approach by comparing with expert opinion on data about 4 terror groups. Joint work with Francesca Spezzano and Aaron Mannes.
Biography:
VS Subrahmanian is a Professor in the Department of Computer Science, Director of the Center for Digital International Government (CDIG), and Co-Director of the Laboratory for Computational Cultural Dynamics (LCCD) at the University of Maryland. From 2004 to 2010, he was Director of the Institute for Advanced Computer Studies (UMIACS) at the University of Maryland. He received the NSF National Young Investigator Award in 1993 and the Distinguished Young Scientist Award from the Maryland Science Center/Maryland Academy of Science in 1997. His research is at the intersection of databases, artificial intelligence, and optimization methods with applications to tracking, monitoring and forecasting behaviors of terrorist groups, socio-cultural groups, global health care, and other areas relevant to most human beings. His work in AI spans rule-based expert systems and logic programs, nonmonotonic reasoning, probabilistic reasoning, temporal reasoning, hybrid reasoning, and software agents. His work in databases focuses on heterogeneous database integration and interoperability, logic databases, probabilistic databases, and multimedia databases. In the last few years, he has been studying how to reason about massive collections of multilingual document collections and mine them for sentiment/opinion information as well as how to mine ontologies directly from text. He has applied his work to the study of foreign cultures and terrorist groups with a view to automatically extracting data about a group's organization and activities and mining this information in order to build stochastic behavioral models of the group which, in turn, can be used to come up with forecasts of future behavior of the group.
Though terrorist networks have been extensively studied, there has been limited work on destabilizing them. We look at the problem of removing "k" people from a terror network (e.g. by capturing them, buying them off, relocating them) in order to minimize the operational effectiveness of the network. We solve this via 3 steps: (i) we develop a framework to predict who succeeds a terrorist when he is ``removed''. (ii) When a set of terrorists is removed, we show how to infer a probability distribution on the resulting possible new networks. (iii) Using the solutions to the previous two problems, we find a set of "k" people to remove from a terror network so that the expected effectiveness of the resulting network is minimized. We test our solutions to the first problem using (i) a synthetic data set and (ii) data sets on terror networks associated with 4 terror groups. For the third problem, we test our approach by comparing with expert opinion on data about 4 terror groups. Joint work with Francesca Spezzano and Aaron Mannes.
Biography:
VS Subrahmanian is a Professor in the Department of Computer Science, Director of the Center for Digital International Government (CDIG), and Co-Director of the Laboratory for Computational Cultural Dynamics (LCCD) at the University of Maryland. From 2004 to 2010, he was Director of the Institute for Advanced Computer Studies (UMIACS) at the University of Maryland. He received the NSF National Young Investigator Award in 1993 and the Distinguished Young Scientist Award from the Maryland Science Center/Maryland Academy of Science in 1997. His research is at the intersection of databases, artificial intelligence, and optimization methods with applications to tracking, monitoring and forecasting behaviors of terrorist groups, socio-cultural groups, global health care, and other areas relevant to most human beings. His work in AI spans rule-based expert systems and logic programs, nonmonotonic reasoning, probabilistic reasoning, temporal reasoning, hybrid reasoning, and software agents. His work in databases focuses on heterogeneous database integration and interoperability, logic databases, probabilistic databases, and multimedia databases. In the last few years, he has been studying how to reason about massive collections of multilingual document collections and mine them for sentiment/opinion information as well as how to mine ontologies directly from text. He has applied his work to the study of foreign cultures and terrorist groups with a view to automatically extracting data about a group's organization and activities and mining this information in order to build stochastic behavioral models of the group which, in turn, can be used to come up with forecasts of future behavior of the group.
Links
Practical information
- General public
- Free
- This event is internal
Contact
- Host : Christoph Koch