Basic  

Overview

Common neighbors refer to the nodes that have edges to two given nodes. Calculation of common neighbors is also used in the Jaccard Similarity algorithm.

Basic Concept

Neighbors

Neighbors are nodes that are connected with node A by edges, and node A itself is not included (suppose A has self-loop edge). Common neighbors are nodes that have edges connected with both node A and B, and are not node A or B. This definition is consistent with the definition of common neighbors in the Jaccard Similarity algorithm: Include neither duplicated neighbors nor nodes to be compared.

In the graph above, the number of common neighbors of the red and green nodes is 2.

Special Case

Lonely Node, Disconnected Graph

Lonely node does not have any neighbor node, the number of common neighbor between lonely node and any node is 0.

For two nodes belong to different connected components, the number of their common neighbor must be 0.

Self-loop Edge

Self-loop edge of a node does not increase the number of neighbors of the node.

Directed Edge

For directed edges, Common Neighbors algorithm ignores the direction of edges but calculates them as undirected edges.

Results and Statistics

Take the graph below as an example, run the Common Neighbor algorithm:

Algorithm results: Calculate A's common neighbors with B, C and D, return node1, node2, num or return node1, node2, ids according to the result type

Algorithm statistics: N/A

Command and Configuration

• Command: algo(common_neighbours)
• Configurations for the parameter params():

Algorithm Execution

Task Writeback

1. File Writeback

Example: Pairing node UUID = 1 and UUID = 2,3,4, calculate common neighbors of each pair, write the algorithm results back to file named cn

algo(common_neighbours).params({
  uuids: 1,
  uuids2: [2,3,4],
  type: 2
}).write({
  file:{
    filename: "cn"
  }
})

2. Property Writeback

Not supported by this algorithm.

3. Statistics Writeback

This algorithm has no statistics.

Direct Return

Example: Pairing node UUID = 1 and UUID = 2,3,4, calculate common neighbors of each pair, define algorithm results (common neighbors list) as alias named neighbors, and return the results

algo(common_neighbours).params({
  uuids: 1,
  uuids2: [2,3,4],
  type: 2
}) as neighbors
return neighbors

Example: Pairing node UUID = 1,2 and UUID = 3,4, calculate common neighbors of each pair, define algorithm results (number of common neighbors) as alias named count, and return the results

algo(common_neighbours).params({
  uuids: [1,2],
  uuids2: [3,4]
}) as count
return count

Streaming Return

Example: Pairing node UUID = 1 and UUID = 2,3,4, calculate common neighbors of each pair, define algorithm results as alias named count, and return the results which have more than 0 common neighbors

algo(common_neighbours).params({
  uuids: 1,
  uuids2: [2,3,4]
}).stream() as count
where count.num > 0
return count

Real-time Statistics

This algorithm has no statistics.