Head's Up! These forums are read-only. All users and content have migrated. Please join us at community.neo4j.com.
03-03-2022 01:48 AM
Error by using Pearson Similarity algorithm.
Indicates 3 arguments, but there are 2 ('params' and 'rel.weight')
MATCH(knop:Compound{prep:'Control'})-[rel:HAS]->(params)
WITH knop, gds.alpha.similarity.asVector(params, rel.weight) AS cVector
MATCH(p1h:Compound)-[rel:HAS]-> (params) WHERE p1h <> knop
WITH knop, p1h, cVector, gds.alpha.similarity.asVector(params, rel.weight) AS p1hVector
RETURN knop.prep AS from, p1h.prep AS to, gds.alpha.similarity.asVector(cVector, p1hVector, {vectorType:'maps'}) AS similarity
ORDER BY similarity DESC
03-03-2022 02:38 PM
You use of asVector in your return statement has three parameters. One is even a map. Is that a valid parameter.
03-04-2022 01:57 AM
I modified the query and got Null
03-04-2022 07:37 PM
You must not have a weight property defined on your ‘compound’ nodes.
03-04-2022 07:43 PM
In your first example, I think the issue is on line 5, were you should be calling pearson method, not asVector method.
All the sessions of the conference are now available online