Comment by culi

19 days ago

There was a post on here about a project in stylometry that analyzed HN users comment history. The tool helped find accounts that had an extremely similar writing style to a given account. The site was soon removed due to privacy concerns but many users with multiple account attested to its accuracy

https://news.ycombinator.com/item?id=33755016

It turns out stylometry is actually a pretty well-developed field. It makes me wanna write an AI browser assistant that can take my comments and stylize them randomly to make it harder to use these sorts of forensics against me

>It makes me wanna write an AI browser assistant that can take my comments and stylize them randomly to make it harder to use these sorts of forensics against me

The old trick years ago was to translate from English to different language and back (possibly repeating). I'd be curious how helpful it is against stylometry detection?

  • The old trick years ago was to translate from English to different language and back (possibly repeating). I'd be curious how helpful it is against stylometry detection?

    If you want to be grouped with foreigners who don't know English, it might work well, although word choices may still be distinctive enough to differentiate even when translated.

    • Assuming the source language is English, going to a romance language and back wouldn't be too hard grammar wise, but could easily wipe out a lot of non-Latin-descended words if you use the right approach to translation.

      1 reply →

On the one side it's a shame this tool was removed because it's very interesting, but on the other hand, the main use case would likely abuse and (cyber)stalking.

That said, best to assume that the various government agencies have tools like this, and better - if you're trying to hide your identity online, don't just change users or go through VPNS/proxies/TOR but change your writing style too.

(Also I'm convinced most VPNs/ proxies / TOR nodes / public access points are honeypots)

A while back the government claimed it had used stylometry to identify Satoshi Nakamoto.

I remember using one of these tools and it falsely identified some other account as being mine. Of course, I only have just this account.