News for hackers. Stuff that matters.

A comment on What it takes to build great machine learning products . Parent comment

Re:What it takes to build great machine learning products by robrenaud on Monday April 16, 2012 @02:26PM (#3848604)
Naive Bayes is a linear classifier, and it makes much stronger assumptions than other linear classifiers.
[ Parent | on HN ]
Re:What it takes to build great machine learning products by irahul on Monday April 16, 2012 @03:55PM (#3849047)
My bad.

Regarding stronger assumptions, is there anything other than independence(thus the name naive) that it assumes?

[ Parent | on HN ]
Re:What it takes to build great machine learning products by tensor on Monday April 16, 2012 @04:18PM (#3849182)
That's the main assumption that people care about. In contrast, logistic regression (aka maximum entropy) does not make this assumption. As a go-to first classifier I would suggest multi-class logistic regression with regularization.
[ Parent | on HN ]