- cons: - If the proposal (m of n) is the majority function
we could infer this function in a unique decision tree.
- pro: - If we have
(experiments) we can fail to
construct the exact function.
- cons: - This method supposes many attributes.
- pro: OK, is a restriction. The best case is with independent
(and meaningfull .-) attributes,
but this method can be used also with additional
attributes that are a new representation of originals.
We must avoid attribute exhaustion.
- cons: - This method is biased toward additive functions.
- pro: - Some functions express an additive condition of some laws in nature.
This work is derived from
Luis Josť Cearra Zabala
Tue Oct 23 11:16:31 CEST 2001