Hacker News Reader

The immediate flaw in that approach is that "patio11 et al" are top contributers because they post a lot and have valuable opinions - NOT because they are always right. They're not (I'm confident they would agree here). Therefore their opinions would be inappropriate as an AI training set.

View on HN

That's true. But doesn't that kind of sidestep the question?

It's always going to be a matter of opinion what constitutes middlebrow-dismissalness, so if we accept that then the question is really:

"Is it possible to automatically determine if this post would be judged in this way by this person (or group of people)?"

It seems like that's a question that has at least a possibility of a "yes" answer.

Sure, totally agree, but a consensus by verified posters was more of what I was aiming for

I guess I am going to pitch my potential solution again:

http://gkosev.blogspot.com/2012/08/fixing-hacker-news-mathem...

it would probably have more weight when i release the test code and datasets though...

Reply on news.ycombinator.com