Apply Bayes Theorem to Optimize IVR Authentication Process

TL;DR

This work addresses IVR authentication security by replacing static credential checklists with a Bayesian risk-aware framework that updates the likelihood of caller fraud after each credential pass/fail. It employs Bayes' Theorem and the Law of Total Probability to compute posterior fraud probabilities and to guide adaptive credential ordering, aiming to maximize deterrence while minimizing legitimate user friction. The study demonstrates that certain credential combinations can drastically reduce posterior fraud risk, and that adaptive sequencing based on evidence can further improve detection without excessive user burden. The approach offers a practical path to robust, risk-based IVR authentication suitable for dynamic fraud landscapes.

Abstract

This paper introduces a Bayesian approach to improve Interactive Voice Response (IVR) authentication processes used by financial institutions. Traditional IVR systems authenticate users through a static sequence of credentials, assuming uniform effectiveness among them. However, fraudsters exploit this predictability, selectively bypassing strong credentials. This study applies Bayes' Theorem and conditional probability modeling to evaluate fraud risk dynamically and adapt credential verification paths.

Figures (3)

• Figure 1: IVR System Flow Diagram
• Figure 2: Correlation Between Credentials
• Figure 3: Sequential IVR Authentication Flow