[gurgeous]

Also see the gigantic map - https://iconmap.io

The blog post is the analysis of the data set, the map is the visualization.

[oehpr]

I wonder if there might be a way to map all these using t-SNE to discrete grid locations? Maybe even an autoencoder. I'd love to see what features it could pick out.

I don't see their data set though. hmmm.

maybe I'll just have to crawl it on my own if I want to do it.

[lgvld]

You can use t-SNE (or even better: UMAP or one of its variation) to create a 2D points cloud, and then use something like RasterFairy [1] to map 2D positions to the cells a grid. It usually works well.

[1] https://github.com/Quasimondo/RasterFairy

[yboris]

side note: instead of t-SNE consider UMAP - provides better results (and it's much faster) https://github.com/lmcinnes/umap

[isoprophlex]

Is the dataset available for download? I couldn't immediately find a download to the dataset in the linked article.

My hands itch to do some dimension reduction on that data and make some nice plots

[nkriege]

We'd be happy to share the data. Reach us at help at gurge.com if you're interested.

[wiz21c]

damn I was thinking about that too :-)

[svdr]

I see a lot of repetitions in the map?

[gurgeous]

It's one icon per domain. Try hovering (on desktop) and you'll see that many domains have the same favicon.

[true_religion]

It also works on mobile if you tap the fav icon.