I am begginer in machine learning. can you provide me with some links to understand how boosting and random forest work and what is pros and cons of these techniques ? also I am interested in which criterias have to use to make model validation and verification after abovementioned models or other machine learning models (except test error).
thank in advance.
Thats depends on the type of data.
https://analyticstraining.com/studentstories-boosting-vs-random-forest/
https://towardsdatascience.com/decision-tree-ensembles-bagging-and-boosting-266a8ba60fd9
https://stackoverflow.com/questions/46190046/gradient-boosting-vs-random-forest
Hi, Levani,
A recently-discovered problem with boosting and RF is that both methods find models in random data. Here are two brief open-access articles on the subject (and a solution):
- https://odajournal.com/2019/03/21/some-machine-learning-algorithms-find-relationships-between-variables-when-none-exist-cta-doesnt
- https://odajournal.com/2019/04/11/effect-of-sample-size-on-discovery-of-relationships-in-random-data-by-classification-algorithms
Thats depends on the type of data.
https://analyticstraining.com/studentstories-boosting-vs-random-forest/
https://towardsdatascience.com/decision-tree-ensembles-bagging-and-boosting-266a8ba60fd9
https://stackoverflow.com/questions/46190046/gradient-boosting-vs-random-forest
Some times the learnin process became unstable. Furthermore, the samplig data is problematic
Thank everyone. Mr Paul Yarnold can I conclude that CTA can always correctly find out if there are really relationship between variables?
- Badges
- Science topic