HomeResearchAwarded Projects Search

Dynamic Statistical Network Informatics

PI: Kathleen Carley, Carnegie Mellon University

Year Selected for Award: 2015

Dynamic Statistical Network Informatics

Principal Investigator: Kathleen Carley, Carnegie Mellon University

Co-Investigator: Joel H. Levine

Years of Award: 2015 - 2020

Managing Service Agency: Office of Naval Research

Project Description:

Current techniques commonly employed by the DoD cannot place heterogeneous data into a homogenous space that allows for scrutinizing large volumes of data, or the most effective use of machine learning and visualization. Graph learning could provide this by finding a 'best-fit' graph to a collection of data. This problem and the need for such techniques has grown with the increase in digital data and the availability of more dynamic data. Our objective is to develop a set of novel scalable techniques for inferring graphs that fit complex data, and are more interpretable than standard social networks. A second objective is to test these across diverse data sets in order to characterize the strength, limitations, and scope for the sub-approaches.

A combination of dynamic network analytic techniques are combined with optimization and machine learning routines to create novel techniques for dimensionality reduction on complex data, that preserves and brings to the forefront any underlying network structure.  Sub-techniques include those for converting attribute data to networks, reducing high dimensional networks to low dimensional networks, sparsifying dense networks, and clustering high dimensional networks.  Multiple diverse data sets, with both known and unknown structures are used for method testing.  Usability of the methods are tested.  Application domains include social media data, archival data, historical data sets and malware for diverse regions around the world.

Select Publications

  • Campedelli, Gian Maria, Mihovil Bartulovic, and Kathleen M. Carley. 2019. Pairwise similarity of jihadist groups in target and weapon transitions. Journal of Computational Social Science pp 1–26.
  • Cruikshank, Iain and Kathleen M. Carley. 2019 – forthcoming. Socio-cultural Cognitive Mapping to Identify Communities and Latent Networks. Lecture Notes in Social Networks.
  • Morgan, Geoffrey & Kathleen M. Carley. 2019 - forthcoming. Characterizing Organizational Micro-Climates in Structural Groups. In Proceedings of the International Conference SBP-BRiMS 2019, Halil Bisgin, Ayaz Hyder, Chris Dancy, and Robert Thomson (Eds.) July 9-12, 2019, Springer.
  • Altman, Neal. Kathleen M. Carley, and Jeffrey Reminga. 2018. ORA User's Guide 2018. Carnegie Mellon University School of Computer Science, Institute for Software Research, Technical Report CMU-ISR-18-103.
  • Eletreby, Rashad & Yong Zhuang, Kathleen M. Carley, and Osman Yagan. 2018. On the Evolution of Spreading Processes in Complex Networks. Carnegie Mellon University School of Computer Science, Department of Electrical and Computer Engineering & Institute for Software Research.
  • Altman, Neal, Kathleen M. Carley, and Jeffrey Reminga. 2017. ORA Datasets 2017. Carnegie Mellon University, School of Computer Science, Institute for Software Research, CMU-ISR-17-112.
  • Altman, Neal, Kathleen M. Carley, and Jeffrey Reminga. 2017. ORA User's Guide 2017. Carnegie Mellon University, School of Computer Science, Institute for Software Research, Technical Report CMU-ISR-17-100.
  • Carley, Kathleen M, Geoffrey Morgan, and Joel H. Levine. 2017. Socio-Cultural Cognitive Mapping. Carnegie Mellon University, School of Computer Science, Institute for Software Research, Technical Report CMU-ISR-17-115.
  • Morgan, Geoffrey, Joel H. Levine, and Kathleen M. Carley. 2017. Socio-Cultural Cognitive Mapping. In Proceedings of the 2017 SBP-BRiMS Conference on Social Computing, Behavioral-Cultural Modeling and Prediction and Behavior Representation in Modeling and Simulation, Washington DC, July 5-8, 2017.
  • Levine, Joel H., and Kathleen M. Carley. 2016. SCM System. Carnegie Mellon University, School of Computer Science, Institute for Software Research, Technical Report CMU-ISR-16-108.