SimGNN

Abstract

Graph similarity search is among the most important graph-based applications, e.g. finding the chemical compounds that are most similar to a query compound. Graph similarity/distance computation, such as Graph Edit Distance (GED) and Maximum Common Subgraph (MCS), is the core operation of graph similarity search and many other applications, but very costly to compute in practice. Inspired by the recent success of neural network approaches to several graph applications, such as node or graph classification, we propose a novel neural network based approach to address this classic yet challenging graph problem, aiming to alleviate the computational burden while preserving a good performance. The proposed approach, called SimGNN, combines two strategies. First, we design a learnable embedding function that maps every graph into an embedding vector, which provides a global summary of a graph. A novel attention mechanism is proposed to emphasize the important nodes with respect to a specific similarity metric. Second, we design a pairwise node comparison method to supplement the graph-level embeddings with fine-grained node-level information. Our model achieves better generalization on unseen graphs, and in the worst case runs in quadratic time with respect to the number of nodes in two graphs. Taking GED computation as an example, experimental results on three real graph datasets demonstrate the effectiveness and efficiency of our approach. Specifically, our model achieves smaller error rate and great time reduction compared against a series of baselines, including several approximation algorithms on GED computation, and many existing graph neural network based models. Our study suggests SimGNN provides a new direction for future research on graph similarity computation and graph similarity search 1 .

Practical Deep-Learning Representation for Fast Heterogeneous Catalyst Screening

“…In addition, we used the attention mechanism, widely used in natural language process , and graph similarity search field, which demonstrated its effectiveness in detecting local information correlated to the global feature. We used the attention algorithm proposed by Bai et al that computes the atomic contribution with respect to the global feature, which is used to find node features close to the global feature. The proposed attention algorithm is applied to the crystal graph after it has passed all convolutional layers.…”

AI-Based Semantic Multimedia Indexing and Retrieval for Social Media on Smartphones

“…Storing these data structures in databases such as Neo4J [45] is typical. Graph embedding can represent graphs mathematically [46] and AI-based approaches optimize pattern matching in graphs [47]. Semantic analysis based on graphs provides comprehensive query languages like SPARQL [48], RDF Schemas and many algorithms for graph transformations.…”

“…In [49], a comparison of algorithms for graph similarity is given. AI-based methods are becoming more common [47]. Here, an efficient graph-based indexing and pattern matching is important, particularly on smartphone hardware.…”

“…Thus, f GC has to be applied once per graph to calculate the graph code, and once the GC is calculated, no more graph traversal is required. Compared to mathematical flattening functions based on linear algebra such as those in [44,47], we follow a more algorithmic approach to define the transformation of a graph into the two-dimensional pixel space based on a colored valuation matrix. An example of this approach is given below in Section 3.5.…”

Computerized Algorithmic Approaches for Evaluating Systems Thinking of Both Engineers and Non-Engineers

“…An implementation of the graph edit distance comparison method was created using Pythonbased code that can input diagrams created in Lucidchart (lucidchart.com) and determine the GED between two diagrams. A similarity rating is then determined from the GED using the normalization described by Bai et al [18]. A comparison was then made between the GED evaluation and the similarity rating determined by the course instructor.…”