r

Scraping 'most read' stories from the BBC

I have had a long-time fascination with the BBC website’s top ten most read table. The most trivial of stories appear to usurp critical world events, spending hours (or days) at number one.

R

R course(s).

Wrangling freedive watch data in R

Towards the end of 2021, I finally plucked up the courage to take a freediving course. At the time, I had never even tried Scuba diving, but a love of swimming and a childish love of trying to swim pool lengths underwater, combined with several inspiring videos, was enough for me to give it a try.

Misrepresentation in maps: mini-survey results from the Ministry of Defence

A couple of years ago, Reka Solymosi and I began a side-project on different ways of visualising spatial data. We were (well, still are) interested in how people interpret maps, and how these interpretations might differ depending the type of map being used, even when the underlying data is the same.

Akmedoids R package for generating directionally-homogeneous clusters of longitudinal data sets

Paper and documentation for an R package to implement anchored k-medoids, a method for clustering longitudinal data.

Update: lockdown crime trends

Over the past few weeks I have spent a bit of time exploring police recorded crime trends before and after the UK-wide lockdown. There has been talk of lockdowns representing the largest criminological experiment in history.