News for hackers. Stuff that matters.

A comment on A machine learning approach to identifying thought markers of suicidal subjects. Parent comment

Re:A machine learning approach to identifying thought marker... by denzil_correa on Tuesday November 08, 2016 @06:58AM (#12899520)
I'm not even sure why people report ONLY accuracy as an evaluation metric. It means nothing - appropriately weighted F scores, precision, recall give more insight into the system.
[ Parent | on HN ]
Re:A machine learning approach to identifying thought marker... by dasboth on Tuesday November 08, 2016 @08:18AM (#12899947)
Completely agree. Although if your binary classes are perfectly balanced then accuracy is a valid metric, but that's rarely the case.

I guess the problem is that an ROC curve wouldn't create flashy headlines.

[ Parent | on HN ]
Re:A machine learning approach to identifying thought marker... by denzil_correa on Tuesday November 08, 2016 @09:31AM (#12900477)
> I guess the problem is that an ROC curve wouldn't create flashy headlines.

You can, you just need to put in a little bit of work. We don't need to specify the exact accuracy in the title.

[ Parent | on HN ]