But, not to be worried, there's fun to be had - with the schedule. I mean, who can read any of this stuff, really:
Tables, Schmables!!! Let's but it into a graph!
Start with a model!
From the tables above, I distilled the following model:
It's actually pretty rich:
- Floors, days and times are connected to eachother by "in-graph-index" relationships.
- Rooms and talks are there
- Talks are part of a track
- Persons can act as speakers and as track hosts.
There's some fun to be had there.
Importing it into Neo4j
Naturally, I started with a spreadsheet. I needed to do some copy/pasting and cleaning of the data, and that's what I use that for. But once it was there, generating the Cypher to create the graph was trivial. Here's the create script to do that yourself - just clone it if you want.
Once we had that, we could start doing some querying on the graph. I have put some sample queries over here.
Here are some interesting ones exploring the graph:
Here we go looking at some talks and tracks and how they are connected to eachother:
And this is probably my favourite: looking at the connectivity between the unfollowable Mark Needham and Yan Cui of Gamesys:
Of course there's plenty more to explore. That's why my friend and colleague Michael Hunger was kind enough to put the database (in read-only mode) on one of his servers - you can take a look at it over here.
Hope you found this useful - see you next week at the conference!
Cheers
Rik
No comments:
Post a Comment