1
$\begingroup$

I have googled this for some time with no luck. All i get are tutorials or articles explaining the classic algorithms like linear regression, random forest, etc.

I would like to know which are the most succesful recent algorithms at this moment for classification with an implementation in Python. I'm interested specially on ensemble methods, such like XG Boost, Catboost and Light Gradient Boosting Machines. Are there any other algorithms similar to these 3 mentioned above with proven sucess capabilities? I would appreciate any help from experienced people that have tried any.

I know there is not such a thing like the best algorithm for anything, what i want to know is if there are any other algorithms that perform the same or better than the ones mentioned above in general. The goal is to solve classification problems using tabular data and supervised learning.

Thanks in advance

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

Deep Learning is your friend. Neural Nets with the growth of publicly available data, and also privately, have beaten every other architecture with the sufficient hardware for it to run properly on.Here's a depiction of why NN's are so popular

Deep Learning

Improve this answer
$\endgroup$
2
  • 1
    $\begingroup$ Hi @Blenz, thank you very much for your answer :). I knew already about dee learning but my problem is not that complex. Using neural nets to solve it may be an overkill. I was interested in algorithms for classic machine learning that can compete or outperform XG Boost, and the others mentioned in the question. $\endgroup$
    – Miguel 2488
    Jul 29, 2019 at 9:33
  • $\begingroup$ yes got it. this could serve as info for beginners who aren't familiar with machine learning concepts $\endgroup$
    – Blenz
    Jul 29, 2019 at 9:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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