Sweeteners could accelerate cognitive decline (neurology.org)

I think this is an interesting place to find unique articles and news. Sometimes for an insightful or well written post. At times for a laugh or good joke. And sometimes to discover some hot new (or new to me) product, or amazing creator, or good person.

I also find a lot of the articles boring, and not interesting, nor worth a read. This happens often. Ditto the threads. I can’t classify my interests into categories or keywords, so RSS is wrong.

Does anyone else feel this way? This is already a high signal place, how do you increase that and maybe tune it specifically for you?

Comments (8)

PaulHoule · 9h ago

I turned this research

https://ontology2.com/essays/ClassifyingHackerNewsArticles/

into a smart RSS reader that does the same for about 110 RSS feeds. Something missing out of most UI's is a "see something once why see it again" interface where you can x something out and not see it again rather than scan and scan and scan and scan.

yorwba · 4h ago

Did you ever manage to improve your classifier beyond what you described in that post or is getting past ≈0.75 AUC ROC simply too difficult.

PaulHoule · 1h ago

For recommendation you can't do that much better because the oracle is fickle. TikTok claims to get 0.83 or so. A better recommender practically has more objectives like diversity. If your recommender shows you just arXiv articles about recommeNeeds pretty soon articles about crypto or soccer or oncology start looking good! I have some ideas of what 'better' is but can't cook them down to one number to optimize.

Part of it is that 'show me an article I'll like' isn't the right question at this point in time when I like almost everything that it shows me, particularly in times like this one when I'm taking photos and developing them and playing Arknights a lot so the ratio of articles it shows me to what it ingests is low. I've been told I'd do better thinking about it from the viewpoint of 'show me the best articles you can given a limited number you can show me' and the right data set for that would be a pairwise comparison, yet the thumbs up/thumbs down task is one that's natural to do for putting articles into a pipeline that I save and/or post to my socials.

I am thinking of how to build a general text classifier which is more state of the art but I'll need to make a test suite of hard but fair problems.

wonger_ · 8h ago

- asking questions in threads where the knowledgeable people are at

- I like to browse /newest and /show, and if something deserves more activity, I'll post a comment. Maybe it will get to the frontpage and attract better discussions

- search hn.algolia.com when I stumble upon interesting links, because maybe there was a past discussion with good commentary

- if there were no past submissions, then submit the link myself to attract commentary

- browsing posts from years ago if I'm bored

- don't get addicted

duxup · 9h ago

I find I get more out of HN making this my go to page:

https://news.ycombinator.com/active

keepamovin · 7h ago

I didn’t know about that! Thank you :-)

fsflover · 9h ago

https://news.ycombinator.com/item?id=16442888

PaulHoule · 9h ago

I've thought about strats to improve my votes/post ratio, one option is to seek out articles that have "gotten more than N votes more than M times" where N might be 200 and M might be 5 and blend them into my mixture.