An Iterative Global Structure-Assisted Labeled Network Aligner

Abstract

Integrating data from heterogeneous sources is often modeled as merging graphs. Given two or more "compatible", but not-isomorphic graphs, the first step is to identify a graph alignment, where a potentially partial mapping of vertices between two graphs is computed. A significant portion of the literature on this problem only takes the global structure of the input graphs into account. Only more recent ones additionally use vertex and edge attributes to achieve a more accurate alignment. However, these methods are not designed to scale to map large graphs arising in many modern applications. We propose a new iterative graph aligner, gsaNA, that uses the global structure of the graphs to significantly reduce the problem size and align large graphs with a minimal loss of information. Concretely, we show that our proposed technique is highly flexible, can be used to achieve higher recall, and it is orders of magnitudes faster than the current state of the art techniques.

SiNA: A Scalable Iterative Network Aligner

“…We presented SINA, a scalable iterative graph aligner. SINA is a careful, shared memory parallelization of a recent sequential graph aligner GSANA [27]. SINA can achieve up to 28.5× speedup on a 32-core machine, reduces the total execution time of a graph alignment problem with 2M vertices and 100M edges from 4.5 hours to under 10 minutes, while retaining the state of the art alignment recall obtained by GSANA.…”

“…SINA's sequential execution times are about 33 times faster than those reported in [27]. Experimental results show that our proposed framework, SINA, reaches up to 28× speedup.…”

“…DBLP (2014-2017): As discussed in GSANA [27], we downloaded two consecutive years of DBLP graphs from 2014 [7] to 2017 [22]. The ground-truth is created using key attribute of author elements.…”

“…More recent ones exploit additional information about vertex and edges for improved alignment, when such information is available. One recent approach is implemented in GSANA [27] which uses global global structure of the graphs to reduce the problem size. GSANA has been shown to have higher recall than the state of the art algorithms while being orders of magnitudes faster.…”

GRASP: Graph Alignment through Spectral Signatures

“…FINAL [35] aligns nodes based on similarity of topology and attributes. GSANA [34] lets pairwise distances to seed nodes guide the matching. Another variant matches weigthed matrices using their spectra [32]; that is inapplicable to the unweighted case.…”