What question did this study set out to answer?

The aim is to improve trust in AI applications in finance by addressing explainability.

December 1, 2020Open Access

Towards Responsible AI for Financial Transactions

View Full Paper

Ask AI

Bookmark

The aim is to improve trust in AI applications in finance by addressing explainability.
Developed a deep neural network for classifying financial transactions using varied inputs.
Used feature importance analysis via SHAP for explanation.
Tested model robustness against targeted evasion attacks.
The SHAP analysis effectively highlighted key features influencing model decisions.
Hybrid clustering and decision tree approach provided clear classification insights.
Model robustness demonstrated resilience against evasion attacks.

No takes yet. Share an insight, caveat, or question.

Empirical study examines explainability in AI financial classification systems, enhancing trust.

Maree et al. (2020) studied this question.

View Full Paper

Ask AI

Bookmark