There are 6 groups of features in the dataset as everyone knows and I’ve been always thinking there should be some reasons behind that.
Followings are the avenues I’ve tried to explore so far:
-
Train models on a feature group (i.e. Dexterity only, Strength only etc), a combination of feature groups (i.e. Intelligence & Strength, Dexterity & Charisma & Constitution etc). There can be so many variations. Take subsets from each group and combine them, ensemble predictions etc etc.
-
Generate some representative features from each feature group (e.g. PCAs, correlations, stds…) and use them for predictions.
-
Limit interactions within a feature group and look at interactions with other feature groups only. (I’ve tried a XGBoost version below.)
https://xgboost.readthedocs.io/en/latest/tutorials/feature_interaction_constraint.html
None of these attempts have led to any meaningful improvement in the metrics so far, unfortunately…
Has anyone tried something similar? Will be great if you could possibly share you 2 cents. Thanks!