The Representation of Meaningful Precision, and Accuracy

TL;DR

This paper critiques traditional precision and accuracy measures, proposing a minimalist rough framework for more meaningful and domain-relevant knowledge representation applicable across various contexts.

Contribution

It introduces a novel compositional knowledge representation approach within a general rough framework to better capture meaningfulness and relevance in models.

Findings

Traditional measures are limited for understanding model relevance.

A new rough framework offers a more meaningful representation.

Potential applicability across diverse problem domains.

Abstract

The concepts of precision, and accuracy are domain and problem dependent. The simplified numeric hard and soft measures used in the fields of statistical learning, many types of machine learning, and binary or multiclass classification problems are known to be of limited use for understanding the meaningfulness of models or their relevance. Arguably, they are neither of patterns nor proofs. Further, there are no good measures or representations for analogous concepts in the cognition domain. In this research, the key issues are reflected upon, and a compositional knowledge representation approach in a minimalist general rough framework is proposed for the problem contexts. The latter is general enough to cover most application contexts, and may be applicable in the light of improved computational tools available.