Optimizing in-count searches with neo4j-CodePudding

I am building a social network like graph where we periodically would like to find all the nodes with an incoming count of greater than 2. E.g.

A->B

C->B

C->A

D->A

E->A

Should return node A.

I am new to neo4j and am still a bit confused about whether they will search through all the available rows. Is there an index I can use to optimize these queries? My search query is like this

EXPLAIN MATCH ()-[r:F]->(b:Person)
WITH b, count(r) AS count
WHERE count > 1
RETURN collect(b.username)

Thank you

CodePudding user response：

Neo4j has a special relationship count store that holds the number of relationships a node has. It allows to get the count of relationships without having to expand them.

You can access relationship count store values using the size() function.

MATCH (b:Person)
WITH b, size((b)<--()) as size
WHERE size > 2
RETURN collect(b.username)