# Consistency of Selection Strategies for Fraud Detection

arXiv: https://arxiv.org/abs/2509.18739 · HTML: https://mlsanity.com/paper/2509.18739/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 34 references that are in the map.
- First posted on arXiv 23 Sep 2025.

## Builds on (most cited first, 5 of 5)

- [A Survey on Practical Applications of Multi-Armed and Contextual Bandits](https://mlsanity.com/paper/1904.10040/) (2019)
- [Deep Contextual Multi-armed Bandits](https://mlsanity.com/paper/1807.09809/) (2018)
- [Social network analytics for supervised fraud detection in insurance](https://mlsanity.com/paper/2009.08313/) (2020)
- [robROSE: A robust approach for dealing with imbalanced data in fraud detection](https://mlsanity.com/paper/2003.11915/) (2020)
- [Unsupervised anomaly detection for discrete sequence healthcare data](https://mlsanity.com/paper/2007.10098/) (2020)

## Abstract (arXiv)

This paper studies how insurers can chose which claims to investigate for fraud. Given a prediction model, typically only claims with the highest predicted propability of being fraudulent are investigated. We argue that this can lead to inconsistent learning and propose a randomized alternative. More generally, we draw a parallel with the multi-arm bandit literature and argue that, in the presence of selection, the obtained observations are not iid. Hence, dependence on past observations should be accounted for when updating parameter estimates. We formalize selection in a binary regression framework and show that model updating and maximum-likelihood estimation can be implemented as if claims were investigated at random. Then, we define consistency of selection strategies and conjecture sufficient conditions for consistency. Our simulations suggest that the often-used selection strategy can be inconsistent while the proposed randomized alternative is consistent. Finally, we compare our randomized selection strategy with Thompson sampling, a standard multi-arm bandit heuristic. Our simulations suggest that the latter can be inefficient in learning low fraud probabilities.