Errata

WITH "https://github.com/neo4j-graph-analytics/book/raw/master/data"
WITH base + "transport-nodes.csv" AS url
LOAD CSV WITH HEADERS FROM url AS row
MERGE (place:Place {id:row.id})
SET place.latitude = toFloat(row.latitude),
place.longitude = toFloat(row.latitude),
place.population = toInteger(row.population)

place.longitude = toFloat(row.latitude) should be place.longitude = toFloat(row.longitude)

WITH "https://github.com/neo4j-graph-analytics/book/raw/master/data"
WITH base + "transport-nodes.csv" AS url
LOAD CSV WITH HEADERS FROM url AS row
MERGE (place:Place {id:row.id})
SET place.latitude = toFloat(row.latitude),
place.longitude = toFloat(row.latitude),
place.population = toInteger(row.population)

place.longitude = toFloat(row.latitude) should be place.longitude = toFloat(row.longitude)

and

WITH base + "transport-nodes.csv" AS url should be
WITH base + "/transport-nodes.csv" AS url

At Closeness Spark code following line is cut
paths_type = ArrayType(
StructType([StructField("id", StringType()), StructField("distance",

Meanwhile at github code* is:
paths_type = ArrayType(StructType([
StructField("id", StringType()),
StructField("distance", IntegerType())
]))

* https://github.com/neo4j-graph-analytics/book/blob/master/scripts/centrality/closeness/cc.py

On page 93 Pivotal nodes are introduced. "if you remove a pivotal node, the new shortest path for the original node pairs will be longer and more costly". However, in Figure 5-7, the third graph marks as pivotal a node that can be removed without increasing the length of the shortest path for A-B.

Figure 5-7 presents a small graph and highlights in blue the pivotal nodes between two nodes A and B. The page indicates that a node is considered pivotal if it lies on *every* shortest path between two nodes, but in each case there is an alternate route with the minimum number of intermediate nodes (3) through the upper node connected to B. Therefore, the rightmost blue node ought to be grey in each of the three graphs.