返回与另一个节点关系最密切的节点

Question

First of all i'm new to neo4j and cypher, so if this is trivial i'm sorry :)

I'm facing a problem I cannot solve, but I will try to explain what I have to do.

I have a graph containing some customers, that have made some orders containing products. This means that the graph is setup like this:

(c:Customer)-[r:PURCHASED]->(o:Order)-[f:ORDERS]->(p:Product)

Here is a representation of the graph I have to work on (This is only an example and not the actual graph):

I have to find customer C, and all the products he has ordered (p1-p6), which is not a problem. What I need to do is find the customer who have ordered most of the same products, and output that customer and the products. This means that customer c2, has ordered 3 of the same products as customer c, compared to c3 who has only ordered 2 of the same. I want c2 then.

Here is what I have so far

MATCH (c:Customer {customerID: "C"})-[r:PURCHASED]->(o:Order)-
[f:ORDERS]->(p:Product),
(p)<-[f1:ORDERS]-(o1:Order)<-[r1:PURCHASED]-(c1:Customer)
WITH c1 AS c1, count(p) AS count, p AS p
WHERE count > 4
RETURN c1,p

This does not really give me what I want. Since when I put count > 4, I get 1 customer and 1 product. This product has 3 f1:ORDERS relationships, to the customer. I wanted a customer (or more customers) that had bought more than 4 of the same products as the customer with ID "C".

Obviously i'm doing something wrong, and I can't seem to figure out how to get the correct result. I hope someone is able to help me

Answer 1

This should work for you:

MATCH
  (c:Customer { customerID: "C" })-[r:PURCHASED]->(o:Order)-[f:ORDERS]->(p:Product),
  (p)<-[f1:ORDERS]-(o1:Order)<-[r1:PURCHASED]-(c1:Customer)
WITH c1, COLLECT(DISTINCT p) AS ps
RETURN c1, ps
ORDER BY LENGTH(ps) DESC 
LIMIT 1;

This query aggregates, for each c1 customer, the distinct products that that customer bought that are in common with the products bought by c . It then sorts those customers (in descending order) by how many products are in common; and returns the top customer, along his collection of shared products.

Here is a console that shows the results with your sample data.

Answer 2

You can use ORDER BY and LIMIT to do what you want:

MATCH (c:Customer {customerID: "C"})-[r:PURCHASED]->(o:Order)-
[f:ORDERS]->(p:Product),
(p)<-[f1:ORDERS]-(o1:Order)<-[r1:PURCHASED]-(c1:Customer)
WITH c1, count(p) AS count, p
ORDER BY count
RETURN c1, p LIMIT 1

This will order the result by count property, and return only the first one, so the one with the most products related.

If this is not what you need (I did not understand your question very well) say it n the comments, I'll update my answer.