Live data from Hacker News

Show HN: Using stylometry to find HN users with alternate accounts

stylometry.net

51–60 of 562 posts

Re: Show HN: Using stylometry to find HN users with alternate accounts

#52
Wow. This gives a lot of false positives, but it found all ~10 of my old accounts over the years.

The most interesting thing is that my writing style changed pretty drastically since a decade ago. Searching for my oldest account matches my earliest usernames, whereas searching this account matched the rest.

The details of the algorithm are fascinating: https://stylometry.net/about Mostly because of how simple it is. I assumed it would measure word embeddings against a trained ML model, but nothing so fancy.

Re: Show HN: Using stylometry to find HN users with alternate accounts

#56
post #44

This is an evil website. We won’t have any anonymity soon. The highest match is my years old banned account that I forgot about. Where did you get the data from?

HN has an Algolia-based API. It’s also very easy to crawl.

I wouldn’t call this evil, however: it’s merely demonstrating a technique that you should be aware of, if you’re a privacy-conscious person. It looks like they also provide some resources for avoiding stylometric detection.

Re: Show HN: Using stylometry to find HN users with alternate accounts

#57
ive had maybe a hundred throwaway accounts on HN over the past ten years. generally, i make an account, say something that is apparently wildly offensive to someone else, get flagged and down-voted and then muted or hell-banned. then i make another account because i never did anything wrong and start the process over again. ive emailed the admins, tried to reason with the admins, it never does any good. the power is held by power-users who flag people -- most of the power of an admin at the end of the day but without any of the accountability. as long as they are following the mainstream dogma, its all good.

anyway, this app was able to identify a lot of my accounts. but a lot of the matches werent me. bold matches were almost all me. but i know there are many more matches than those that were listed. it mainly showed my most recent accounts.

i think most people would get a sick feeling in their stomach if they tried this app. i dont think people are prepared for a world where you can type someones name into an app like this and produce everything ever recorded online that was created by that person. not only this but everything highlighted and summarized to answer any question about that person. this is what advanced ai will bring us. an information implosion where the planet-sized ocean of data that is just floating all around us suddenly and violently coalesces into the objects of our new societal calculus. violent is a good word. and this is just the change that one can see coming with ai.

Re: Show HN: Using stylometry to find HN users with alternate accounts

#58

Wow. This gives a lot of false positives, but it found all ~10 of my old accounts over the years. The most interesting thing is that my writing style changed pretty drastically since a decade ago. Searching for my oldest account matches my earliest usernames, whereas searching this account matched the rest. The details of the algorithm are fascinating: https://stylometry.net/about Mostly because of how simple it is.…

It works like a charm for me too.

I put in my username and found my pre-echelon alt, possibilistic.

(Echelon was taken when I registered possibilistic, but it must have been unused and dropped.)

Re: Show HN: Using stylometry to find HN users with alternate accounts

#59

Wow. This gives a lot of false positives, but it found all ~10 of my old accounts over the years. The most interesting thing is that my writing style changed pretty drastically since a decade ago. Searching for my oldest account matches my earliest usernames, whereas searching this account matched the rest. The details of the algorithm are fascinating: https://stylometry.net/about Mostly because of how simple it is.…

Yeah top 20 is a little excessive because in my own tests I found that top 20 is only marginally more accurate than top 10. You can get a more academic explanation [here](https://www.tandfonline.com/doi/abs/10.1080/09296174.2011.53...). I was amazed too because it seemed too easy!
Post reply on HN