Real-Time Data Quality Indicators for Monitoring Predictive Power Degradation in Machine Learning Credit Scoring Models: An Extended DAMA Framework Perspective
Abstract
Index (CSI), Kolmogorov improve statistical rigor, we recommend reporting uncertainty for discriminatory metrics using bootstrap confidence intervals and calibrating thresholds through baseline sampling rather than fixed rules.The manuscript demonstrates the monitoring workflow using transparent synthetic aggregates to illustrate numerical traceability without exposing proprietary data, and it provides an escalation playbook for triage, investigation, and model action. 2.
Fields of Study & Subjects
Financial Distress and Bankruptcy Prediction
Data Quality and Management
Imbalanced Data Classification Techniques
Publication details
- Journal / source
- Arab Journal for Scientific Publishing
- Volume
- 8
- Issue
- 88
- Publication date
- 3 February 2026