Does allshortestpath in gds can not outcome the links and nodes for each path?

[lingxuan9]

Hi, I have a question about allshortestpath algorithm in gds.

When I run the query like this:

CALL gds.alpha.allShortestPaths.stream({
nodeQuery: 'MATCH (n:Loc) RETURN id(n) AS id',
relationshipQuery: 'MATCH (n:Loc)-[r:ROAD]-(p:Loc) RETURN id(n) AS source, id(p) AS target, r.cost AS cost',
relationshipWeightProperty: 'cost'
})
YIELD sourceNodeId, targetNodeId, distance
WITH sourceNodeId, targetNodeId, distance
WHERE gds.util.isFinite(distance) = true

MATCH (source:Loc) WHERE id(source) = sourceNodeId
MATCH (target:Loc) WHERE id(target) = targetNodeId
WITH source, target, distance WHERE source <> target

RETURN source.name AS source, target.name AS target, distance
ORDER BY distance DESC, source ASC, target ASC
LIMIT 10

I find that when I call the gds.alpha.allShortestPaths, the output is only sourceNodeId, targetNodeId and distance. Is it true? I want get the information of links and nodes for each path. Can I get them by calling gds?

Many thanks for your help.

[filantrop]

It is not possible with that algorithm.
It all depends on your usecase.

Do you surely need all shortest paths for all pair of nodes?

[lingxuan9]

Thanks for your reply.
Yes, I think so. Since I need to calculate some complex network indicators according to the information of links between each two node

[lingxuan9]

I don't need all shortest paths for all pair of nodes, but for each node pair, I need the detail path of the shortest path, like which nodes this path across.

[filantrop]

I think you can find some different solutions here:

All shortest paths between a set of nodes Cypher

For that approach (shortestPath from each of the starting nodes to the closest node within a pre-matched set) you can use APOC path expander procedures. We have the concept of endNodes, where you can supply a list of possible nodes that you want to expand to and end upon, and we can also supply a limit for the number of paths to return per procedure call. We'll use apoc.path.spanningTree() here, which uses a type of traversal uniqueness such that a single node is never visited more than once, an…

[lingxuan9]

Thank you for your help.
Yes, according to this post, I have similar scenario with tenkaleakshay94, but in the case first. That is, I need the shortest path to every label2 node.
So now, I have a large network and I need to get the shortest path information for all node pairs.
But it is too much time consuming.
So I am looking for methods can improve the performance.
Do you have any ideas about how to improve the query performance?

[filantrop]

Without knowing you real usecase I'm not sure I can help you with the performance problem.

When running an algorithm to get all nodepairs shortest paths will take alot of time to perform. Because it will traverse all nodes multiple times.

You mentioned earlier that you need to calculate on nodes that is crossed. Perhaps a centrality algorithm would help you.