How to select Certain Wavelet lvl for feature extraction?

[Mojo-jo]

I liked your detailed explanation regarding the wavelet and its implementation in python. I have been through several articles relating wavelet to be used as feature extraction, however, in some articles, they mentioned that they select certain detail/approximate coefficient level where they implement feature extraction for that lvl. I want to ask how is it possible to decide which level to choose for a feature extraction?

[taspinar]

@Mojo-jo I think it comes with a lot of domain expertise or frequent use of wavelets. Of course you can also calculate which frequency range each level corresponds with and choose a decomposition level for which you reasonably expect the behaviour to exhibit.
But, what maybe is more practical is that you choose the maximum decomposition level an by keeping track of from which decomposition level each feature comes from, and calculating the feature importance of each feature, you can see how important each decomposition level. If there are decomposition levels which turn out to have insignificant feature importance then you can leave out the features generated from that level.