Hi everyone,
In machine learning, suppose we have a training database consisting of chronological ordered instances, from least recent to most recent, e.g. a user's behavior on social media.
Knowing we want to predict the user's future behavior according to the past behavior, is it more appropriate in this case to use the hold-out validation (the least recent set for training and the more recent for testing) instead of the k-fold cross-validation?
Thanks in advance.
Dear Researcher,
You have to pay attention to the fact that a time series(chronological) is not randomly distributed data. Do something like forward chaining where you’ll be able to model on past data then look at forward-facing data. Do folding like this
fold 1 : training [1], test [2]
fold 2 : training [1 2], test [3]
fold 3 : training [1 2 3], test [4]
fold 4 : training [1 2 3 4], test [5]
fold 5 : training [1 2 3 4 5], test [6] like so on
Regards
Suheel
Hi,
In this case the best way is to use one-out-level cross-validation.
Regards;
Dear Researcher,
You have to pay attention to the fact that a time series(chronological) is not randomly distributed data. Do something like forward chaining where you’ll be able to model on past data then look at forward-facing data. Do folding like this
fold 1 : training [1], test [2]
fold 2 : training [1 2], test [3]
fold 3 : training [1 2 3], test [4]
fold 4 : training [1 2 3 4], test [5]
fold 5 : training [1 2 3 4 5], test [6] like so on
Regards
Suheel
- Badges
- Science topic
- Similar topics
- Social Science
- Media
- Social Media