Examining Clandestine Social Networks for the Presence of Non-Random Structure
Abstract
This thesis develops a tractable, statistically sound hypothesis testing framework for the detection, characterization, and estimation of non-random structure in clandestine social networks. Network structure is studied via an observed adjacency matrix, which is assumed to be subject to sampling variability. The vertex set of the network is partitioned into k mutually exclusive and collectively exhaustive subsets, based on available exogenous nodal attribute information. The proposed hypothesis testing framework is employed to statistically quantify a given partition's relativity in explaining the variability in the observed adjacency matrix relative to what can be explained by chance. As a result, valuable insight into the true structure of the network can be obtained. Those partitions that are found to be statistically significant are then used as a basis for estimating the probability that a relationship tie exists between any two vertices in the complete vertex set of the network. The proposed methodology aids in the reduction of the amount of data required for a given network, focusing analyses on those attributes that are most promising. Ample effort is given to both model demonstration and application, including an example using open-source data, illustrating the potential use for the defense community and others.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2007
- Accession Number
- ADA466683
Entities
People
- Joshua S. Seder
Organizations
- Air Force Institute of Technology