CAPE: Automatically Predicting Changes in Group Behavior

  • Amy SlivaEmail author
  • V.S. Subrahmanian
  • Vanina Martinez
  • Gerardo Simari


There is now intense interest in the problem of forecasting what a group will do in the future. Past work [1, 2, 3] has built complex models of a group’s behavior and used this to predict what the group might do in the future. However, almost all past work assumes that the group will not change its past behavior. Whether the group is a group of investors, or a political party, or a terror group, there is much interest in when and how the group will change its behavior. In this paper, we develop an architecture and algorithms called CAPE to forecast the conditions under which a group will change its behavior. We have tested CAPE on social science data about the behaviors of seven terrorist groups and show that CAPE is highly accurate in its predictions—at least in this limited setting.


Hide Markov Model Action Variable Group Behavior Terrorist Group Change Table 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Martinez, V., Simari, G.I., Sliva, A., Subrahmanian, V.S.: Convex: Similarity-based algorithms for forecasting group behavior. IEEE Intelligent Systems 23(4) (2008) 51–57CrossRefGoogle Scholar
  2. 2.
    Subrahmanian, V.S., Albanese, M., Martinez, V., Nau, D., Reforgiato, D., Simari, G.I., Sliva, A., Wilkenfeld, J.: Cara: A cultural adversarial reasoning architecture. IEEE Intelligent Systems 22(2) (2007) 12–16CrossRefGoogle Scholar
  3. 3.
    Subrahmanian, V.S.: Cultural modeling in real-time. Science 317(5844) (2007) 1509–1510CrossRefGoogle Scholar
  4. 4.
    Schrodt, P.: Forecasting conflict in the balkans using hidden markov models. In: Proc. American Political Science Association meetings. (2000)Google Scholar
  5. 5.
    Bond, J., Petroff, V., O’Brien, S., Bond, D.: Forecasting turmoil in indonesia: An application of hidden markov models. In: International Studies Association Convention, Montreal. (2004) 17–21Google Scholar
  6. 6.
    Martinez, V., Simari, G.I., Sliva, A., Subrahmanian, V.S.: The SOMA terror organization portal (STOP): Social network and analytic tools for the real-time analysis of terror groups. In Liu, H., Salerno, J., Young, M., eds.: Social Computing, Behavioral Modeling and Prediction, Spring Verlag (2008) 9–18Google Scholar
  7. 7.
    Khuller, S., Martinez, V., Nau, D., Simari, G.I., Sliva, A., Subrahmanian, V.S.: Finding most probable worlds of probabilistic logic programs. In: Proc. 2007 International Conference on Scalable Uncertainty Management. Volume 4772., Springer Verlag Lecture Notes in Computer Science (2007) 45–59Google Scholar
  8. 8.
    Khuller, S., Martinez, V., Nau, D., Simari, G., Sliva, A., Subrahmanian, V.S.: Computing most probable worlds of action probabilistic logic programs: Scalable estimation for 1030,000 worlds. Annals of Mathematics and Artificial Intelligence 51(2-4) (2007) 295–331zbMATHCrossRefMathSciNetGoogle Scholar
  9. 9.
    Wilkenfeld, J., Asal, V., Johnson, C., Pate, A., Michael, M.: The use of violence by ethnopolitical organizations in the middle east. Technical report, National Consortium for the Study of Terrorism and Responses to Terrorism (2007)Google Scholar
  10. 10.
    Bowerman, B., O’Connell, R., Koehler, A.: Forecasting, Time Series and Regression. Southwestern College Publ (2004)Google Scholar
  11. 11.
    Ullman, J.: Principles of Database and Knowledge Base Systems. Volume 2. Computer Science Press, Maryland (1989)Google Scholar

Copyright information

© Springer-Verlag/Wien 2009

Authors and Affiliations

  • Amy Sliva
    • 1
    Email author
  • V.S. Subrahmanian
    • 1
  • Vanina Martinez
    • 1
  • Gerardo Simari
    • 1
  1. 1.University of Maryland College ParkCollege ParkUSA

Personalised recommendations