CODEWITHSUNDEEP
neo4jcypher
Stetco Oana
Stetco Oana

Reputation: 125

Load nodes and create relations from a single csv file

I have to load a CSV file having an id (cui), and a name of a disease, some of the ids are repeating themselves but they have a slightly different name. I would like to create the nodes with unique id and nodes for all the other names which are slightly different. the nodes with alternative name will have a relationship [:HAS_ALTERNATIVE_NAME] with the initial nodes.

I have a query which is appending the alternative names in the node property, this time I will like to normalize the problem by creating a node for each alternative name.

USING PERIODIC COMMIT 
LOAD CSV WITH HEADERS FROM 'file:///Concepts_50000.csv' AS line 
FIELDTERMINATOR '\t'
MERGE (d:Disease{id: line.CUI})
ON CREATE SET 
    d.prefered_name = line.name,
    d.alternative_name = line.name
ON MATCH SET
    d.alternative_name = d.alternative_name+', '+line.name;

In the end, I will like to have something like this situation, one disease having alternative names nodes.

Ideal result of the querry

Upvotes: 0

Views: 44

Answers (1)

cybersam
cybersam

Reputation: 67044

This may do what you want:

USING PERIODIC COMMIT 
LOAD CSV WITH HEADERS FROM 'file:///input.csv' AS line 
FIELDTERMINATOR '\t'
MERGE (d:Disease {id: line.CUI})
MERGE (n:DiseaseName {name: line.name})
MERGE (d)-[: HAS_ALTERNATIVE_NAME]->(n)

[UPDATED]

For better performance, make sure you have indexes on:

  • :Disease(id)
  • :DiseaseName(name)

Upvotes: 1

Related Questions