Neo4j技术问询：已知查询不同边类型语句，如何返回各类型及出现次数？

Hey there! Great question—you're already halfway there with your existing query. To get each distinct relationship type along with how many times it appears, you just need to add an aggregation function and let Neo4j handle the grouping for you.

Here's the adjusted query you need:

MATCH (a)-[r]->(b) 
RETURN type(r) AS relationship_type, COUNT(r) AS occurrence_count 
ORDER BY occurrence_count DESC

Let me break this down for you:

• type(r): This still grabs the name of each relationship type, just like your original query. We alias it as relationship_type to make the result set cleaner.
• COUNT(r): This aggregation function counts every instance of the relationship r. Since we're pairing this with type(r), Neo4j automatically groups all relationships by their type before counting—no extra GROUP BY clause is needed (though you could add GROUP BY type(r) if you prefer explicit syntax, it won't change the result).
• ORDER BY occurrence_count DESC: This is optional, but it's super useful to see the most common relationship types at the top of your results.

For example, if your graph has FOLLOWS and LIKES relationships, your output might look like this:

If you only want to count distinct relationships (though that's rarely needed here, since each r is a unique relationship instance), you could use COUNT(DISTINCT r) instead—but in most cases, plain COUNT(r) is exactly what you want.

内容的提问来源于stack exchange，提问作者Solomon Bothwell