Over the past few years, I've experimented with various forms of visualisations (using D3.js, the GapMinder Motion Chart), and various other technical bits'n'bobs (various Python scripts for example, using the Natural Language Toolkit). I've listed a few examples below. Most experiments will be posted up on Labs, or referenced by the blog.
Early preview of our Visualisation and Chart Tool over at Refined Practice. Using D3.js (and in the future using other technologies) this enables non-technical editors to publish sophisticated interactive visualisations and charts on a variety of web platforms.
During the summer of 2012, I carried out an analysis of 2.5 millions News and Politics search queries for the BBC, as part of their BBC News Labs projects. Taking the data from Experian Hitwise and the BBC's own internal analytics, I visualised the data using the GapMinder Motion Chart). It was powerful analysis, highlighting revealing insights into public opinion. The data was also referenced in this blog post: Are We Interested in Nadine Dorries?