Struc2Vec Walk

✓ File Writeback ✕ Property Writeback ✓ Direct Return ✓ Stream Return ✕ Stats

Overview

The Struc2Vec Walk is a biased random walk. This is one of the crucial componments of the Struc2Vec framework, where the walk is performed in a constructed multilayer weighted graph rather than the original graph. Please refer to the Struc2Vec algorithm for the details.

Syntax

Command：algo(random_walk_struc2vec)
Parameters:

Name	Type	Spec	Default	Optional	Description
ids / uuids	[]`_id` / []`_uuid`	/	/	Yes	ID/UUID of nodes to start random walks; start from all nodes if not set
walk_length	int	≧1	`1`	Yes	Depth of each walk, i.e., the number of nodes to visit
walk_num	int	≧1	`1`	Yes	Number of walks to perform for each specified node
k	int	[1, 10]	/	No	Number of layers of the constructed multilayer weighted graph, which should not exceed the diameter of the original graph
stay_probability	float	(0,1]	/	No	The probability of walking in the current level
limit	int	≧-1	`-1`	Yes	Number of results to return, `-1` to return all results

Example

The example graph is as follows:

File Writeback

Spec	Content	Description
filename	`_id`,`_id`,...	IDs of visited nodes

UQL
algo(random_walk_struc2vec).params({
  walk_length: 5,
  walk_num: 1,
  k: 4,
  stay_probability: 0.8
}).write({
  file:{
    filename: 'walks'
}})

Results: File walks

File
J,G,I,G,
I,H,I,H,G,
H,G,H,I,H,
G,H,I,G,
F,G,H,I,H,
E,C,B,C,
D,C,E,F,G,
C,E,C,A,C,
B,A,D,E,
A,B,A,D,

Direct Return

Alias Ordinal	Type	Description	Columns
0	[]perWalk	Array of UUIDs of visited nodes	`[_uuid, _uuid, ...]`

UQL
algo(random_walk_struc2vec).params({
  ids: ['J'],
  walk_length: 6,
  walk_num: 3,
  k: 4,
  stay_probability: 0.8
}) as walks
return walks

Results: walks

[10, 6, 3, 6]

[10, 7, 6, 5, 6, 5]

[10, 7, 10, 7, 6, 7]

Stream Return

Alias Ordinal	Type	Description	Columns
0	[]perWalk	Array of UUIDs of visited nodes	`[_uuid, _uuid, ...]`

UQL
algo(random_walk_struc2vec).params({
  ids: ['J'],
  walk_length: 6,
  walk_num: 30,
  k: 5,
  stay_probability: 0.7
}).stream() as walks
where size(walks) == 6
return walks

Results: walks

[10, 7, 6, 5, 3, 4]

[10, 7, 8, 7, 6, 5]

[10, 7, 6, 4, 6, 7]