How can I add a condition in "SIMILARITY"

[mehdi_ajroud]

Excuting the query of similarity was successuful by using this :

MATCH (at1:Attrib)<-[r:ASSIGNED]-(c:Contrat)-[x:ASSIGNED]->(at2:Attrib)
MERGE  (at1)-[s:SIMILARITY]-(at2)

And then , to display only Attribs which are working together , I worked with this query and it's fine :

MATCH (at:Attrib)<-[r]-(c:Contrat)
WITH DISTINCT c.id AS Contrat , collect(DISTINCT at.label) AS Attributaires, count( DISTINCT at.id) AS Count
Return  Attributaires;

and this the result :

My question is how can I remove duplicates for "at" in the graph and not only in the list . if 2 at have the same ID and they re displayed then we display only one at .

[mark_needham]

I'm not sure that I understand exactly what you'd like to do...could you maybe share a sample dataset that I can create locally and then say what you'd like the output of your query to be?

[mehdi_ajroud]

Actually this is the graph that I obtained after excuting "similarity" .

And here is the list of "Attributes" which are working together :

What I want to do exactly is : deleting the duplicates nodes which have the same ID (ID : 111 apearing twice for example here in the graph , I want to havejust one instead of two nodes with the same ID .

[mehdi_ajroud · Accepted solution]

Here is the solution :

gist.github.com

https://gist.github.com/jruts/fe782ff2531d509784a24b655ad8ae76

neo4j_delete_duplicate_nodes.md

# How to delete duplicate nodes and their relationships in neo4j with cypher based on a property of that node?
The problem is easy to understand. We have _'duplicate'_ nodes in our database based on the _'id'_ field on the node properties.

Well this sounds easy enough, until you have to actually do it.

## First step
My first attempt was to try and figure out which nodes are actualy duplicate (based on a property on the node).
This seems to be pretty straightforward.

_Cypher_:

This file has been truncated. show original