NumericOverflow
NumericOverflow

Reputation: 1039

How to count distance/hops/length (# relationships) between nodes using cypher?

I'm trying to determine the "distance" between nodes in a graph. Using the example movie database, I would like to return actor nodes with the distance away from Kevin Bacon. Using the image below for effectively what I'm looking for.

How do I build this into my cypher query? This seems possible, I just can't think of a way to do it as my cypher-foo isn't very advanced yet :(

MATCH p=(bacon:Person {name:"Kevin Bacon"})-[*1..5]-(hollywood)
WHERE hollywood.name in (['Helen Hunt', 'Ed Harris'])
RETURN p

FYI - My neo4j database is v4.0

Example Distance metric

Upvotes: 0

Views: 1333

Answers (1)

cybersam
cybersam

Reputation: 67044

This query:

MATCH p=(bacon:Person {name:"Kevin Bacon"})-[*1..5]-(hollywood)
WHERE hollywood.name IN ['Helen Hunt', 'Ed Harris']
UNWIND
  REDUCE(s = [], i IN RANGE(0, LENGTH(p)) |
    s + {dist: i, node: NODES(p)[i]}
  ) AS data
RETURN data.dist AS distance, COLLECT(DISTINCT data.node) AS nodes

returns this result:

╒══════════╤═════════════════════════════════════════════════════════════════════╕
│"distance"│"nodes"                                                              │
╞══════════╪═════════════════════════════════════════════════════════════════════╡
│0         │[{"name":"Kevin Bacon","born":1958}]                                 │
├──────────┼─────────────────────────────────────────────────────────────────────┤
│1         │[{"title":"Apollo 13","tagline":"Houston, we have a problem.","releas│
│          │ed":1995},{"title":"A Few Good Men","tagline":"In the heart of the na│
│          │tion's capital, in a courthouse of the U.S. government, one man will │
│          │stop at nothing to keep his honor, and one will stop at nothing to fi│
│          │nd the truth.","released":1992},{"title":"Frost/Nixon","tagline":"400│
│          │ million people were waiting for the truth.","released":2008}]       │
├──────────┼─────────────────────────────────────────────────────────────────────┤
│2         │[{"name":"Tom Hanks","born":1956},{"name":"Ed Harris","born":1950},{"│
│          │name":"Bill Paxton","born":1955},{"name":"Cuba Gooding Jr.","born":19│
│          │68},{"name":"Jack Nicholson","born":1937},{"name":"Ron Howard","born"│
│          │:1954}]                                                              │
├──────────┼─────────────────────────────────────────────────────────────────────┤
│3         │[{"title":"Cast Away","tagline":"At the edge of the world, his journe│
│          │y begins.","released":2000},{"title":"Twister","tagline":"Don't Breat│
│          │he. Don't Look Back.","released":1996},{"title":"As Good as It Gets",│
│          │"tagline":"A comedy from the heart that goes for the throat.","releas│
│          │ed":1997},{"title":"Apollo 13","tagline":"Houston, we have a problem.│
│          │","released":1995}]                                                  │
├──────────┼─────────────────────────────────────────────────────────────────────┤
│4         │[{"name":"Helen Hunt","born":1963},{"name":"Ed Harris","born":1950}] │
└──────────┴─────────────────────────────────────────────────────────────────────┘

Notice that the movie "Apollo 13" appears in the results at distances of 1 and 3. And that "Ed Harris" appears at distances of 2 and 4. This is because there are (undirected) paths of different lengths from "Kevin Bacon" to those two nodes.

Upvotes: 0

Related Questions