Title
A Controlled Sensing Approach To Graph Classification
Keywords
Complex Networks; Controlled Sensing; Estimation Theory; Graph Classification; Social Networks
Abstract
The problem of classifying graphs with respect to connectivity via partial observations of nodes is posed as a composite hypothesis testing problem with controlled sensing. An observation at a node is a subset of edges incident to the node on the complete graph drawn according to a probability model, which are modeled as conditionally independent given their neighborhoods. Connectivity is measured through average node degree and is classified with respect to a threshold. A simple approximation of the controlled sensing test is derived and simulated on Erdös-Rènyi Model A graphs to characterize error probabilities as a function of expected stopping times. It is shown that the proposed test achieves favorable tradeoffs between the classification error and the number of measurements and further outperforms existing approaches, especially at low target error rates. Furthermore, the proposed test achieves asymptotically optimal error performance, as the error rate goes to zero. © 2013 IEEE.
Publication Date
10-18-2013
Publication Title
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Number of Pages
5573-5577
Document Type
Article; Proceedings Paper
Personal Identifier
scopus
DOI Link
https://doi.org/10.1109/ICASSP.2013.6638730
Copyright Status
Unknown
Socpus ID
84890482974 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/84890482974
STARS Citation
Ligo, Jonathan G.; Atia, George K.; and Veeravalli, Venugopal V., "A Controlled Sensing Approach To Graph Classification" (2013). Scopus Export 2010-2014. 6352.
https://stars.library.ucf.edu/scopus2010/6352