100 Days of ML — Day 4 — What Was The Hulk Doing With A Cluster Algorithm?

In the Oscar-snubbed 2012 film The Avengers, Earth’s Mightest Heroes, built up from five previous films to come together in a Thanksgiving feast of action-driven hero antics, which, like any good Thanksgiving, had shouting, sibling rivalries, and

Near the end of the first act, Mark Ruffalo as Bruce Banner notes that to find Loki, he’ll “rough out a tracking algorithm based on cluster recognition,” once all the spectrometers in the world are spec’d for gamma rays.

Acne. You won’t like me when I have acne.

At first, it sounds like Hollywood blabbery, something some executive producer overheard on an airplane and said, “Hey! That can go in *my* movie. Say that thing again about the chicken nugget alogrithm.”

But what if it was accurate? What if Hollywood actually knew what it was doing for once?

Let’s talk about what clustering is for a moment. Basically, you have data that doesn’t fit easily into any handy dandy x-y relationship. Maybe it looks like this:

Really stepping up your MS Paint game, yeah, Jimmy?

If I told you to split this three ways, intuitively you can probably see the three types of data.

Loki is somewhere in Germany in the smiley part of the world.

The Incredible Hulk’s task was to find the Tessaract and SHIELD’s NSA network of cell phones and cameras wasn’t cutting it, so he opted to “rule out a few places.” Obviously, Loki and/or the Tessaract is leaving behind a gamma ray signature and Bruce Banner is the master of applied gamma rays and has spent some time in the algorithm world of data mining, giving us, the viewer, a fantastic intro the future of AI!

I imagine his algorithm ultimately yielded a graph like this:

“Guys, I think this might be in New York, like every other superhero thing”

Looking at this graph a few times, I think you can pinpoint where the Tessaract might have been. I think we can applaud Hollywood on their efforts for this one. They would’ve been better off changing a few lines to track the iridium that Loki was going to need or maybe even that scepter.

And if Hollywood’s gonna go full techno-babble, they should have talked about K-Means Clustering.

Thanks, Wiki!

Jimmy Murray is a Florida based comedian who studied Marketing and Film before finding himself homeless. Resourceful, he taught himself coding, which led to a ton of opportunities in many field, the most recent of which is coding away his podcast editing. His entrepreneurial skills and love of automation have led to a sheer love of all things related to AI.





Source: Deep Learning on Medium