openml / EvaluationEngine

Sources of the Java Evaluation Engine
8 stars 6 forks source link

meta-features store vector of numbers instead of aggregates #16

Open janvanrijn opened 6 years ago

janvanrijn commented 6 years ago

Currently we store (for numeric columns):

Where X = {mean, stdev, kurtosis, skewness}. Something similar for information theoretic measures of nominal atts.

This selection is arbitrary and not well supported in the literature.

Much better would be to store a vector of each value per attribute, giving the possibility to researchers to calculate these values client-side.