Beyond Black-Box Explanations: Monte Carlo Dropout for Uncertainty-Aware Explainable AI in Marketing Analytics
Bibliographic record
Abstract
Marketing AI systems increasingly rely on explainable artificial intelligence (XAI) to justify customer targeting, yet current methods provide no indication of when explanations can be trusted, creating risks of unreliable targeting and reduced campaign effectiveness. We present a systematic framework for uncertainty-aware XAI using Bayesian neural networks with Monte Carlo Dropout and SHAP analysis to identify when explanations are unreliable. Applied to a bank marketing dataset of 45,211 customers, our Bayesian model achieved an AUC of 0.84 and quantified feature-level reliability u sing 3 0 Monte Carlo samples per prediction. Results show that 23% of explanations exhibit high epistemic uncertainty and should not guide automated decisions; notably, demographic features such as age (uncertainty: 0.0033) and loan status (0.0038) proved less reliable than behavioral and temporal attributes. This framework enables automatic flagging of untrustworthy explanations with quantitative reliability thresholds, equipping marketing teams with uncertainty-aware decision-making tools that act as early warning systems to prevent costly errors in customer acquisition and retention.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.013 | 0.104 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.002 | 0.006 |
| Insufficient payload (model declined to judge) | 0.003 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".