The Optimal Input-Independent Baseline for Binary Classification: The Dutch Draw

TL;DR

This paper proves that the Dutch Draw baseline is the optimal input-independent method for binary classification evaluation under certain conditions, simplifying baseline selection.

Contribution

It establishes the Dutch Draw as the best input-independent baseline for positional-invariant measures in binary classification evaluation.

Findings

Dutch Draw is proven to be the optimal input-independent baseline.

It applies to all positional-invariant measures assuming random shuffling.

Simplifies baseline selection in binary classification evaluation.

Abstract

Before any binary classification model is taken into practice, it is important to validate its performance on a proper test set. Without a frame of reference given by a baseline method, it is impossible to determine if a score is `good' or `bad'. The goal of this paper is to examine all baseline methods that are independent of feature values and determine which model is the `best' and why. By identifying which baseline models are optimal, a crucial selection decision in the evaluation process is simplified. We prove that the recently proposed Dutch Draw baseline is the best input-independent classifier (independent of feature values) for all positional-invariant measures (independent of sequence order) assuming that the samples are randomly shuffled. This means that the Dutch Draw baseline is the optimal baseline under these intuitive requirements and should therefore be used in practice.