| Title: | Algorithmic Framework for Representational Learning on Graphs |
| Version: | 0.1.0 |
| Description: | Given any graph, the 'node2vec' algorithm can learn continuous feature representations for the nodes, which can then be used for various downstream machine learning tasks.The techniques are detailed in the paper "node2vec: Scalable Feature Learning for Networks" by Aditya Grover, Jure Leskovec(2016),available at <doi:10.48550/arXiv.1607.00653>. |
| License: | GPL (≥ 3) |
| Encoding: | UTF-8 |
| LazyData: | true |
| RoxygenNote: | 7.1.0 |
| Imports: | data.table, igraph, word2vec, rlist, dplyr, vctrs, vegan |
| Depends: | R (≥ 2.10) |
| NeedsCompilation: | no |
| Packaged: | 2020-12-27 15:31:28 UTC; Administrator |
| Author: | Yang Tian [aut, cre], Xu Li [aut], Jing Ren [aut] |
| Maintainer: | Yang Tian <tianyang1211@126.com> |
| Repository: | CRAN |
| Date/Publication: | 2021-01-14 09:00:02 UTC |
node2vec: Algorithmic Framework for Representational Learning on Graphs
Description
Given any graph, the 'node2vec' algorithm can learn continuous feature representations for the nodes, which can then be used for various downstream machine learning tasks.The techniques are detailed in the paper "node2vec: Scalable Feature Learning for Networks" by Aditya Grover, Jure Leskovec(2016),available at <arXiv:1607.00653>.
Author(s)
Maintainer: Yang Tian tianyang1211@126.com
Authors:
Xu Li
Jing Ren
6 edges information between two genes of human
Description
A dataset containing the 6 interactions of genes
Usage
gene_edges
Format
A data frame with 6 rows and 2 variables:
- gene1
human genes
- gene2
human genes
Source
Algorithmic Framework for Representational Learning on Graphs
Description
Algorithmic Framework for Representational Learning on Graphs
Usage
node2vecR(
data,
p = NULL,
q = NULL,
directed = NULL,
num_walks = NULL,
walk_length = NULL,
dim = NULL
)
Arguments
data |
input data for edges consisting of at least two columns of data and if there are weights,it must be in the third column. |
p |
return parameter.Default to 1. |
q |
in-out parameter.Default to 1. |
directed |
the network is directed or undirected.Default to undirected. |
num_walks |
number of walks per node.Default to 10. |
walk_length |
number of nodes in each walk.Default to 80. |
dim |
embedding dimensions.Default to 128. |
Value
embedding results for each node
Examples
#Parameters can be customized as needed
data(gene_edges)
use_data<-gene_edges
emb<-node2vecR(use_data,p=2,q=1,num_walks=5,walk_length=5,dim=10)