SMARAGD: Learning SMatch for Accurate and Rapid Approximate Graph Distance

TL;DR

SMARAGD introduces neural network-based methods to approximate graph similarity scores efficiently, enabling large-scale applications by significantly reducing computation time while maintaining accuracy.

Contribution

It proposes novel neural network models that approximate Smatch graph similarity scores in linear or constant time, addressing NP-completeness issues.

Findings

Neural networks can predict Smatch scores with reduced error.

Linear-time approximation via machine translation framework.

Constant-time prediction using Siamese CNN.

Abstract

The similarity of graph structures, such as Meaning Representations (MRs), is often assessed via structural matching algorithms, such as Smatch (Cai and Knight, 2013). However, Smatch involves a combinatorial problem that suffers from NP-completeness, making large-scale applications, e.g., graph clustering or search, infeasible. To alleviate this issue, we learn SMARAGD: Semantic Match for Accurate and Rapid Approximate Graph Distance. We show the potential of neural networks to approximate Smatch scores, i) in linear time using a machine translation framework to predict alignments, or ii) in constant time using a Siamese CNN to directly predict Smatch scores. We show that the approximation error can be substantially reduced through data augmentation and graph anonymization.