Sticky tape and string: Learning faster from TripAdvisor reviews

Last week the British Museum wrote about how they were making sense of their TripAdvisor reviews, using natural language processing techniques and data visualisation — work they presented at MCNx. It was good-looking (if dense) stuff, which surfaced new insights into what visitors found important about their experience, and pointed towards some root causes. But I couldn’t help thinking that the approach suffered from throwing away the ratings data, and was very resource-intensive— the article concludes with plans for two PhD data science interns and a partnership with the Alan Turing Institute to continue the work. This makes it pretty daunting for other museums to join in the fun.

I trained an A.I. to imagine movie titles from ACMI’s collection

My first few weeks at ACMI (Australian Centre for the Moving Image) have been a whirlwind of discovery, exploration, planning and excitement around its forthcoming renewal. One of my projects is to continue to flesh out explorations of machine learning and related AI techniques to ACMI’s collection of Australian moving image works.

This is not that project.