Find high weighted mutal friend in Neo4J Database

I am new in graph database and I am making a fiends relationship. So, here in the screen shot you can see R<->A, R<->B, R<->C (<-> follow each other), now you can see A->D,E,G,H and B->D,F,G,H and C->H,I now you can see the H is common in all these and D and G node common for A and B. So, its a very high possibility that H should be a friend of R and D and G also friend of R. So, could anyone know to write a cypher query to get the H,D and G node.

@nimrod answer is totally correct and here is the final cypher query

MATCH(u:User{name:"R"})-[:FOLLOWS]->(:User)-[:FOLLOWS]->(unew:User)
WITH COUNT(unew) AS totalNew,unew as finalUsers
WHERE totalNew>=2 AND unew.name<>"R"
Return totalNew,finalUsers 

One option is:

MATCH (r{name: 'R'})-->()-->(b)
WITH count(b) as itemsCount, b.name as name
WHERE itemsCount >= 2
RETURN name, itemsCount

Which for the example data:

MERGE (nr:Node {name:"R"})
MERGE (na:Node {name:"A"})
MERGE (nb:Node {name:"B"})
MERGE (nc:Node {name:"C"})
MERGE (nd:Node {name:"D"})
MERGE (ne:Node {name:"E"})
MERGE (nf:Node {name:"F"})
MERGE (ng:Node {name:"G"})
MERGE (nh:Node {name:"H"})
MERGE (ni:Node {name:"I"})


MERGE (nr)-[:FOLLOWS]->(na)
MERGE (nr)-[:FOLLOWS]->(nb)
MERGE (nr)-[:FOLLOWS]->(nc)
MERGE (na)-[:FOLLOWS]->(nd)
MERGE (na)-[:FOLLOWS]->(ne)
MERGE (na)-[:FOLLOWS]->(ng)
MERGE (na)-[:FOLLOWS]->(nh)
MERGE (nb)-[:FOLLOWS]->(nd)
MERGE (nb)-[:FOLLOWS]->(nf)
MERGE (nb)-[:FOLLOWS]->(ng)
MERGE (nb)-[:FOLLOWS]->(nh)
MERGE (nc)-[:FOLLOWS]->(nh)
MERGE (nc)-[:FOLLOWS]->(ni)

returns:

╒════╤══════════╕
│name│itemsCount│
╞════╪══════════╡
│"H" │3         │
├────┼──────────┤
│"D" │2         │
├────┼──────────┤
│"G" │2         │
└────┴──────────┘