Risk, Trust, and Bias: Causal Regulators of Biometric-Enabled Decision Support

TL;DR

This paper introduces a comprehensive taxonomy of risk, trust, and bias as causal regulators in biometric decision support systems, linking their assessment to causal inference for improved decision-making.

Contribution

It presents a novel taxonomy that models the causal relationships among R-T-B factors and demonstrates practical assessment methods in biometric and epidemiological contexts.

Findings

Causal relationships among R-T-B factors are crucial for system performance.

Practical experiments assess trust in synthetic biometric data.

The approach extends to decision support in epidemiological surveillance.

Abstract

Biometrics and biometric-enabled decision support systems (DSS) have become a mandatory part of complex dynamic systems such as security checkpoints, personal health monitoring systems, autonomous robots, and epidemiological surveillance. Risk, trust, and bias (R-T-B) are emerging measures of performance of such systems. The existing studies on the R-T-B impact on system performance mostly ignore the complementary nature of R-T-B and their causal relationships, for instance, risk of trust, risk of bias, and risk of trust over biases. This paper offers a complete taxonomy of the R-T-B causal performance regulators for the biometric-enabled DSS. The proposed novel taxonomy links the R-T-B assessment to the causal inference mechanism for reasoning in decision making. Practical details of the R-T-B assessment in the DSS are demonstrated using the experiments of assessing the trust in synthetic biometric and the risk of bias in face biometrics. The paper also outlines the emerging applications of the proposed approach beyond biometrics, including decision support for epidemiological surveillance such as for COVID-19 pandemics.