Neo4j / Cypher语言 - 返回语句中的条件计数

Question

目前我得到了以下Cypher查询:

MATCH (topic:TopicNode)-[r]-(article:ArticleNode)-[b]->(author:AuthorNode)
MATCH (topic)-[v]->(source:SourceNode)
WHERE  r.relevancy > 0.002
RETURN topic, count(DISTINCT article), count(DISTINCT author), count(DISTINCT source)

工作良好，并为每个主题提供预期的结果。

现在我想获得以下附加信息:每个主题具有积极情绪(例如article.sentiment > 0)和消极情绪的文章的数量。

是否有方法将条件语句添加到计数操作符?

例如:

RETURN count(article.sentiment > 0)

因为这似乎不适合我。

Answer 1

您可以尝试使用列表推导式:

MATCH (topic:TopicNode)-[r]-(article:ArticleNode)-[b]->(author:AuthorNode) 
MATCH (topic)-[v]->(source:SourceNode) 
WHERE  r.relevancy > 0.002 
WITH topic, count(DISTINCT article) AS articlesCount, count(DISTINCT author) AS authorsCount, count(DISTINCT source) AS sourceCount, collect(DISTINCT article) AS articles
RETURN topic, articlesCount, authorsCount, sourceCount, size([x IN articles WHERE x.sentiment > 0 | x ]) AS positiveSentiment, size([x IN articles WHERE x.sentiment < 0 | x ]) AS negativeSentiment 

在这里，我们收集列表中的文章，然后使用列表推导，过滤出具有积极和消极情绪的文章，并计算它们的计数。试一试。