Fast Sequence-Based Embedding with Diffusion Graphs
B. Rozemberczki, and R. Sarkar. (2020)cite arxiv:2001.07463Comment: Source code available at: https://github.com/benedekrozemberczki/diff2vec.
Abstract
A graph embedding is a representation of graph vertices in a low-dimensional
space, which approximately preserves properties such as distances between
nodes. Vertex sequence-based embedding procedures use features extracted from
linear sequences of nodes to create embeddings using a neural network. In this
paper, we propose diffusion graphs as a method to rapidly generate vertex
sequences for network embedding. Its computational efficiency is superior to
previous methods due to simpler sequence generation, and it produces more
accurate results. In experiments, we found that the performance relative to
other methods improves with increasing edge density in the graph. In a
community detection task, clustering nodes in the embedding space produces
better results compared to other sequence-based embedding methods.
Description
[2001.07463] Fast Sequence-Based Embedding with Diffusion Graphs
%0 Generic
%1 rozemberczki2020sequencebased
%A Rozemberczki, Benedek
%A Sarkar, Rik
%D 2020
%K diff2vec embedding graph
%T Fast Sequence-Based Embedding with Diffusion Graphs
%U http://arxiv.org/abs/2001.07463
%X A graph embedding is a representation of graph vertices in a low-dimensional
space, which approximately preserves properties such as distances between
nodes. Vertex sequence-based embedding procedures use features extracted from
linear sequences of nodes to create embeddings using a neural network. In this
paper, we propose diffusion graphs as a method to rapidly generate vertex
sequences for network embedding. Its computational efficiency is superior to
previous methods due to simpler sequence generation, and it produces more
accurate results. In experiments, we found that the performance relative to
other methods improves with increasing edge density in the graph. In a
community detection task, clustering nodes in the embedding space produces
better results compared to other sequence-based embedding methods.
@misc{rozemberczki2020sequencebased,
abstract = {A graph embedding is a representation of graph vertices in a low-dimensional
space, which approximately preserves properties such as distances between
nodes. Vertex sequence-based embedding procedures use features extracted from
linear sequences of nodes to create embeddings using a neural network. In this
paper, we propose diffusion graphs as a method to rapidly generate vertex
sequences for network embedding. Its computational efficiency is superior to
previous methods due to simpler sequence generation, and it produces more
accurate results. In experiments, we found that the performance relative to
other methods improves with increasing edge density in the graph. In a
community detection task, clustering nodes in the embedding space produces
better results compared to other sequence-based embedding methods.},
added-at = {2020-12-11T17:04:29.000+0100},
author = {Rozemberczki, Benedek and Sarkar, Rik},
biburl = {https://www.bibsonomy.org/bibtex/2e940570a9383c66d0584cc57515c49ba/parismic},
description = {[2001.07463] Fast Sequence-Based Embedding with Diffusion Graphs},
interhash = {3680b3216eb37d91a153550048962141},
intrahash = {e940570a9383c66d0584cc57515c49ba},
keywords = {diff2vec embedding graph},
note = {cite arxiv:2001.07463Comment: Source code available at: https://github.com/benedekrozemberczki/diff2vec},
timestamp = {2020-12-11T17:04:29.000+0100},
title = {Fast Sequence-Based Embedding with Diffusion Graphs},
url = {http://arxiv.org/abs/2001.07463},
year = 2020
}