Graph pipeline development
In the last chapter, we learned how to interface Python with a Neo4j database. We harnessed Neo4j’s long-term graph storage solution to set up a more realistic, production-like system, which could be queried to find air travel routes, according to several parameters.
We set the graph database up by writing a large amount of data to it using static queries. At the point of setting up the database, we knew what data we wanted it to hold, and we wrote Cypher queries to get our data in, in bulk. The focus of the resulting graph was on delivering read query results to, for example, a frontend web application. These read queries were executed against the database only when we had a question to ask of the data.
However, in reality, these graph database systems often serve as the backend to applications that frequently and automatically send queries to a graph database. These queries may not only be driven by direct user behavior but also executed to...
