《图算法》中Graph Data Science代替Graph Algorithms (Neo4j)代码（第四章）

Neo4j 4.1将GraphAlgorithms用Graph Data Science Library代替。《图算法》一书中的示例需要改写，本文按照书中的示例，改写了代码，并实际运行，结果与书中一致。

第四章 GDS代码

首先要创建内存数据投影，Graph Data Science使用内存数据库。以下代码中的’myGraph’就是GDS中要求的graphName。

call gds.graph.create("myGraph","Place","EROAD") 
YIELD graphName, nodeCount, relationshipCount;

• Shortest Path with Neo4j

原代码

MATCH (source:Place {id: "Amsterdam"}),
 (destination:Place {id: "London"})
CALL algo.shortestPath.stream(source, destination, null)
YIELD nodeId, cost
RETURN algo.getNodeById(nodeId).id AS place, cost

GDS代码

MATCH (source:Place {id: "Amsterdam"}),(destination:Place {id:"London"})
CALL gds.alpha.shortestPath.stream({nodeProjection: 'Place', 
relationshipProjection: 'EROAD',
startNode: source,
endNode: destination,
relationshipWeightProperty: null})
YIELD nodeId, cost
RETURN gds.util.asNode(nodeId).id AS name, cost

• Shortest Path (Weighted) with Neo4j

原代码

MATCH (source:Place {id: "Amsterdam"}),
 (destination:Place {id: "London"})
CALL algo.shortestPath.stream(source, destination, "distance")
YIELD nodeId, cost
RETURN algo.getNodeById(nodeId).id AS place, cost

GDS代码

MATCH (source:Place {id: "Amsterdam"}),(destination:Place {id: "London"})
CALL gds.alpha.shortestPath.stream({nodeProjection: 'Place', 
relationshipProjection:{
EROAD:{
type: '