Cypher aggregation and collect - neo4j

Question

Given the following result in my cypher query, where people are a collection:
[
{
"people": [
{
"id": 24749,
"matches": 1
},
{
"id": 26026,
"matches": 1
},
{
"id": 26223,
"matches": 1
},
{
"id": 25121,
"matches": 1
},
{
"id": 24632,
"matches": 1
},
{
"id": 25708,
"matches": 1
},
{
"id": 25182,
"matches": 1
},
{
"id": 24826,
"matches": 1
},
{
"id": 26186,
"matches": 1
},
{
"id": 27001,
"matches": 1
},
{
"id": 24243,
"matches": 1
},
{
"id": 27255,
"matches": 1
},
{
"id": 27145,
"matches": 1
},
{
"id": 24126,
"matches": 1
},
{
"id": 27463,
"matches": 1
},
{
"id": 24069,
"matches": 1
},
{
"id": 25210,
"matches": 1
},
{
"id": 24994,
"matches": 1
},
{
"id": 27331,
"matches": 1
},
{
"id": 25793,
"matches": 1
},
{
"id": 27312,
"matches": 1
},
{
"id": 26206,
"matches": 1
},
{
"id": 24252,
"matches": 1
},
{
"id": 24714,
"matches": 2
},
{
"id": 24612,
"matches": 1
},
{
"id": 26964,
"matches": 1
},
{
"id": 27101,
"matches": 1
},
{
"id": 26730,
"matches": 1
},
{
"id": 27211,
"matches": 1
},
{
"id": 24783,
"matches": 2
},
{
"id": 25336,
"matches": 1
},
{
"id": 24128,
"matches": 1
},
{
"id": 26186,
"matches": 1
},
{
"id": 25125,
"matches": 2
},
{
"id": 24069,
"matches": 3
},
{
"id": 24607,
"matches": 1
},
{
"id": 27055,
"matches": 1
},
{
"id": 25336,
"matches": 3
},
{
"id": 24128,
"matches": 2
},
{
"id": 26716,
"matches": 1
},
{
"id": 27331,
"matches": 1
},
{
"id": 24069,
"matches": 1
}
]
}
]
How can I (with cypher) iterate the people collection and find the ones with the same "id", sum the "matches" items together and then add a new item called "duplicates" or similar.
Example result I'm trying to get:
[
{
"people": [
{
"id": 24069,
"matches": 5, // all the "matches" of the duplicate 24069's added together
"duplicates": 3 // how may times the id 24069 was found in the collection called people
},
// etc...
]
}
]

Answer 1

The query:
match (p:people) return distinct(p.id),sum(p.matches)
should probably give you what you need

Answer 2

This is kinda complicated, as the simplest way is to tear your collection apart and reassemble it. It's a lot of steps, but here is the Cypher to do that as well as comments on what each step is doing.
// initial matching (whatever you match on to get your collection)
MATCH (n:People) WITH collect(n) as people
// Tear people collection apart for possessing
UNWIND people as p
// Reduce on id
WITH COLLECT(DISTINCT p.id) as id, p.matches as m
// For each id, sum and count the matches
WITH {id:id, matches:SUM(m), duplicates:COUNT(m)} as people
// Recollect rows into collection
RETURN COLLECT(people) as people