Following on from last weeks post where we looked at how to model Prem Rugby data in Neo4j, we’ll now start to look at how we can extract information and insight.

Continuing on the theme of graph databases, I thought this week I would have a look at how we can model and gain insight from some rugby data through Neo4j. The data source is an Opta SuperScout XML which I’ve parsed but unfortunately can’t share the data source itself. The result is a few different tables including:

• Fixture data - scoreline, referee, team ID’s
• Team data - which players played, shirt numbers, position, number of minutes.
• Match stats - each instance/action from throughout the match assigned to each player and team who performed the action.

Last week we built a Prog Rock knowledge graph using data from Wikidata and MusicBrainz. This week we’ll see how we can use Neo4j, Cypher and some of Neo4j’s Graph Data Science Library to create some insight from the knowledge graph. And as a bonus for coming back and reading this after part one, I’ll spare readers the tenuously linked Prog Rock related subheadings…

Welcome to Round 6 predictions from the Elo rating system. Firstly a quick update on the force-velocity dashboard that I posted about last week.

Recently I’ve been exploring graph databases and their potential applications. After getting setup with Neo4j, I thought I’d learn how to use the graph query language Cypher and as I’m a bit of a Prog Rock nause decided I would create a knowledge graph based on 1970’s UK and Itallian prog bands.

Welcome to my latest post. There are two topics for this one, a double whammy if you will - Prem predictions and Force Velocity Profiling.