Clusters topic guide

[zslade]

Type of PR

• BUG
• FEAT
• MAINT
• DOC

Is your Pull Request linked to an existing Issue or Pull Request?

Give a brief description for the solution you have provided

Note to reviewers: This is a work in progress and may undergo some significant changes in light of developing plots for metrics and doing more thinking about how they should best be used!

Topic guide for evaluating clusters. Consists of:

• An overview motivating the use of graphs and graph metrics to evaluate the results of data linking and discussing what good looks like
• A run through of graph metrics currently available with Splink and how they apply to evaluating linked data
• Instructions on how to compute graph metrics using the new functionality (more details to be added to a future iteration)

At present, there isn't a focussed discussion on the various levers that can be pulled to change clusters (e.g. threshold vs constraints of comparisons vs blocking rules) which might be useful. Could wait for future iteration though.

PR Checklist

• Added documentation for changes
• Added feature to example notebooks or tutorial (if appropriate)
• Added tests (if appropriate)
• Updated CHANGELOG.md (if appropriate)
• Made changes based off the latest version of Splink
• Run the linter

[ADBond]

This looks in good shape, thanks for all your work in this! Think you have done a good job explaining complicated things clearly, think this will be a really good addition to the docs

[RossKen]

Good to go from my side now - @zslade please merge whenever you are happy