An Audit Framework for Technical Assessment of Binary Classifiers

TL;DR

This paper introduces an audit framework with 20 KPIs and a traffic light risk assessment method for evaluating the fairness, transparency, and ethics of binary classifiers like random forest and logistic regression models, aiding regulatory compliance.

Contribution

It presents a novel, comprehensive audit framework with KPIs and a risk assessment method for technical evaluation of binary classifiers, aligned with AI regulations.

Findings

The framework effectively assesses model fairness, transparency, and discrimination.

Open-source dataset experiments demonstrate the framework's practical applicability.

Explainability methods like SHAP are evaluated within the framework.

Abstract

Multilevel models using logistic regression (MLogRM) and random forest models (RFM) are increasingly deployed in industry for the purpose of binary classification. The European Commission's proposed Artificial Intelligence Act (AIA) necessitates, under certain conditions, that application of such models is fair, transparent, and ethical, which consequently implies technical assessment of these models. This paper proposes and demonstrates an audit framework for technical assessment of RFMs and MLogRMs by focussing on model-, discrimination-, and transparency & explainability-related aspects. To measure these aspects 20 KPIs are proposed, which are paired to a traffic light risk assessment method. An open-source dataset is used to train a RFM and a MLogRM model and these KPIs are computed and compared with the traffic lights. The performance of popular explainability methods such as kernel- and tree-SHAP are assessed. The framework is expected to assist regulatory bodies in performing conformity assessments of binary classifiers and also benefits providers and users deploying such AI-systems to comply with the AIA.