References & Citations
Condensed Matter > Disordered Systems and Neural Networks
Title: Systematic identification of statistically significant network measures
(Submitted on 24 Jun 2003 (v1), last revised 27 Jan 2005 (this version, v3))
Abstract: We present a novel graph embedding space (i.e., a set of measures on graphs) for performing statistical analyses of networks. Key improvements over existing approaches include discovery of "motif-hubs" (multiple overlapping significant subgraphs), computational efficiency relative to subgraph census, and flexibility (the method is easily generalizable to weighted and signed graphs). The embedding space is based on {\it scalars}, functionals of the adjacency matrix representing the network. {\it Scalars} are global, involving all nodes; although they can be related to subgraph enumeration, there is not a one-to-one mapping between scalars and subgraphs. Improvements in network randomization and significance testing--we learn the distribution rather than assuming gaussianity--are also presented. The resulting algorithm establishes a systematic approach to the identification of the most significant scalars and suggests machine-learning techniques for network classification.
Submission history
From: Etay Ziv [view email][v1] Tue, 24 Jun 2003 17:48:55 GMT (291kb)
[v2] Wed, 25 Jun 2003 02:04:20 GMT (290kb)
[v3] Thu, 27 Jan 2005 22:01:36 GMT (38kb)
Link back to: arXiv, form interface, contact.