An exploratory statistical analysis of Akira and Ghost in the Shell

Link to paper

At the top of this post, I’ve linked my paper analysing the films Akira and Ghost in the Shell . These are both seminal films that have greatly influenced modern horror/thriller/futuristic films, and as a result, they have been picked over by numerous reviewers and analysts.

However, my approach to analysing these films is quite different from the conventional methods. I use data analysis tools to make observations about “macro” trends in the films, and try to rationalise these observations using scenes and clips. If nothing, you should enjoy the pretty pictures on Page 6 and Page 12, which constitute the bulk of my data visualisation work.

Thoughts on this methodology

The feedback from the use of this method is mixed. It gives reasonably good results when we are dealing with anime films (as it did for the films I chose to look at). I suppose this success can be attributed to the tight control on lighting and colors the artists can exert in such a medium. Dealing with real life environments allows for little creative experimentation, and for most real-life films, color palletes are more akin to side effects of physical circumstances, not intentionally curated creative visions.

This method is much, much more time consuming compared to a standard movie review/analysis, for which the most time consuming task is probably just watching the film a couple times. In comparison, neglecting the actual visualisations, simply processing all the frames of a film takes over 4-5 hours on my 2012 laptop. In a world where speed is king, especially in media, these turnaround times are simply not viable for independent journalists. Perhaps, this method becomes much more viable when you have a large team of data scientists and software engineers at your disposal.

With that said, I still hope my project inspires audiences to reconsider how they’ve looked at the particular films Akira and Ghost in the Shell (both of which, by the way, are criminally underrated on MAL) and perhaps, even reimagine the entire methodology of film analyses with me. With computers as our “bicycles for the mind”, there is definitely a lot left to be done in the film department, and I hope that my analysis functions as a glimpse of that potential.