1
$\begingroup$

I have learned several classifiers in Machine learning - Decision tree, Neural network, SVM, Bayesian classifier, K-NN, Markov process...etc.

Can anyone please help to understand when I should prefer one of the classifier over other - for example - in which situation(nature of data sets, etc) I should prefer decision tree over neural net OR which situation SVM might work better than Bayesian OR what type s of problems are appropriate to apply decision tree. neural net or SVM or bayesian ??

Thanks.

$\endgroup$
  • $\begingroup$ I think you should consider asking this on stats.stackexchange.com, which might yield more attention/better answers. $\endgroup$ – Kirill Jul 12 '13 at 1:42
1
$\begingroup$

The question you might want to ask is whether a linear model (Logistic Regression or Lasso) will explain the data better than a non-linear model (Random Forest). You might also want to consider whether you would like to have a generative model with a distribution over labels (Bayesian models) or a discriminative model (SVM). It's important to consider the number of features or degrees of freedom (Neural Networks contain millions of parameters vs. non-parametric model like K-NN).

In practice, it's common to choose a set of models based on the above criteria and ensemble them together, e.g. in a soft-voting classifier that takes into account predictions of several classifiers.

| cite | improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.