THE RESULT
What the data shows
The largest computed metric is 0.4818 for Logistic regression; the smallest is 0.3158 for Prior baseline. Metric: average_precision (Average precision).
Duration, demographics and financial account fields are excluded. Row order is a coarse chronological proxy.
| model | accuracy | balanced accuracy | macro f1 | roc auc | average precision | brier |
|---|---|---|---|---|---|---|
| Prior baseline | 0.6842 | 0.5 | 0.4062 | 0.5 | 0.3158 | 0.2779 |
| Logistic regression | 0.6806 | 0.5272 | 0.4906 | 0.6632 | 0.4818 | 0.2353 |
| Random forest | 0.683 | 0.5036 | 0.4214 | 0.674 | 0.4641 | 0.2365 |
THE METHOD
From source to answer
Fixed logistic and forest benchmarks against a prior baseline on the final source-order fifth.
The Python source exposes this study’s transformations. The complete project download includes shared preparation and evaluation routines.
EVALUATION
How the result was checked
First 80% of ordered observations train; final 20% test.
Fixed models; default decision threshold; no holdout tuning.
Target and input features
Target: subscribed
Inputs: contact, month, day, campaign, pdays, previous, poutcome
THE NEXT DECISION
What follows from the finding
Evaluate ranking and calibration before considering a consent-aware operational pilot.
Where the conclusion stops
Observational marketing records do not identify campaign uplift. Month and row order are not precise timestamps; repeated-client identifiers are unavailable. Models are research diagnostics, not financial eligibility or automated contact decisions.
Related studies may reuse observations or holdouts. These are historical analyses; associations and backtests do not demonstrate commercial impact. Further model tuning needs new, untouched evaluation data.