Hacker News Reader

Comment by costco | Hacker News Reader

I'm reading a book about electricity markets. Modeling electricity prices using a mean-reverting model with "jumps", exotic derivatives, etc. Pretty exciting stuff. Though you need a million dollars in liquid net worth to trade with any ISO so it's a bit out of my league. Really interesting author too: https://en.wikipedia.org/wiki/Iris_Mack

Am working on some large scale authorship attribution stuff on the side and have an Arty A7 that I want to make do something besides flash purple LEDs but I have some health bullshit that's killing my motivation. But I'm working on it.

View on HN

The market operators always need people who can help make sure the market is operating correctly and that market power is not being abused

- "authorship attribution stuff"

Can you tell more about it? I've done something about it in my master's and PhD, but I'm away from this area for some time and don't know the actual state of the art...

It's like my previous project https://news.ycombinator.com/item?id=33755016 but on a much larger amount of data. Rewrote everything in C++, tried to minimize allocations etc but still have some things to optimize. Current algorithm is just the cosine delta method on word/character {1,2,3}-grams as described in https://academic.oup.com/dsh/article/32/suppl_2/ii4/3865676 combined with a few other features from the Writeprint paper.

I have no academic or professional linguistic background so it's definitely not state of the art, but it works reasonably well and from what I can tell in the literature the fancier methods work slightly better but have significantly worse performance. The beauty of the cosine delta is that I can put the vectorized representations of each author into something like hnswlib (InnerProductSpace) or Milvus and get really fast comparisons.

If you get to the point where you have some insight on trading electricity... i know some people who have some capital..

Reply on news.ycombinator.com