VVibeFootprintWebsite intelligence

VibeFootprint research data · n=81

What an uncertainty band changed across 81 website evaluations

Across 20 repeated Development assignments, a 0.38–0.62 indeterminate band withheld a median 13.6% of decisions. Precision improved among decided cases, but unresolved cases still reduced overall positive recall.

Format
Uncertainty data brief
For
Practitioners interpreting website similarity scores and uncertainty
Reading time
9 minutes

Published by VibeFootprint EditorialPublished · Last reviewed

Dataset and protocol

Twenty repeated Development assignments on 81 usable captures

The starting collection contained 169 attempted captures. Technical acquisition left 81 usable websites: 28 with the historical strong-AI benchmark label and 53 with the stable-human benchmark label. The evaluation used 20 deterministic, class-stratified assignments with five folds each.

Each training fold used deterministic minority oversampling; test folds were not oversampled. The model was logistic with L2 regularization of 10, a 0.5 decision threshold and a predefined inclusive indeterminate range from 0.38 to 0.62. The resulting scores express orientation toward this Development benchmark—not calibrated authorship probability.

Decision matrix

Repeated Development metrics before applying the indeterminate band

The median summarizes the 20 assignments; the P10–P90 range shows variation across those assignments. These are complete-case Development metrics, subject to the acquisition limitation in the 169-attempt audit.

MetricMedianP10–P90Interpretation boundary
Precision76.7%71.0–81.5%Share of predicted positive cases carrying the positive benchmark label
Recall78.6%78.6–82.1%Share of positive benchmark cases predicted positive
Specificity86.8%83.0–90.6%Share of negative benchmark cases predicted negative
Accuracy85.2%81.5–86.4%Share of all usable cases matching their benchmark label
ROC AUC88.2%87.2–90.0%Ranking performance within this Development sample

Decision matrix

What changed when scores from 0.38 through 0.62 were left unresolved

An indeterminate band trades decision coverage for cleaner decided cases. Reporting only the improved decided-case metrics would hide the unresolved part of the sample.

QuantityMedian across assignmentsObserved rangePractical meaning
Decided coverage85.2%82.7–90.1%Most cases remained outside the indeterminate band
Abstention rate13.6%9.9–17.3%Roughly one in seven cases received no binary decision at the median
Precision among decided cases87.5%76.9–91.7%Precision rose after excluding ambiguous scores
Recall among decided positive cases80.8%77.8–84.0%Recall looks stronger when unresolved positives are omitted
Overall positive recall with abstentions unresolved75.0%67.9–78.6%The full decision policy recovered fewer positive cases than the decided-only view suggests

Decision matrix

A deterministic perturbation simulation produced few boundary changes

Two inverse jitters within ±5% were applied to each non-binary source count. This tests arithmetic sensitivity under a defined simulation; it does not reproduce browser or website change over time.

Simulation resultValueUnitCorrect interpretation
Comparisons3,240score comparisonsAll recorded perturbation comparisons in the frozen artifact
Median absolute score change0.32points on a 0–100 scaleHalf of simulated changes were no larger than 0.32 points
P90 absolute score change1.12points on a 0–100 scaleNinety percent were no larger than 1.12 points
Maximum absolute score change5.69points on a 0–100 scaleThe most sensitive recorded simulated comparison
Binary threshold flips0.52%of comparisonsA small share crossed the 0.5 binary threshold
Qualitative band changes1.91%of comparisonsA larger share crossed one of the broader reporting bands

Interpretation

An uncertainty band is a decision policy, not an accuracy upgrade

Ten websites were indeterminate by their mean score. That is useful product information: the visible evidence did not justify a crisp binary orientation under the predefined band. The band raised median precision among decided cases from 76.7% to 87.5%, but that comparison is conditional on leaving some cases unresolved.

The responsible product pattern is to show the unresolved state, preserve continuous evidence and avoid translating the orientation score into authorship certainty. A wider or narrower band would change coverage and error trade-offs and therefore needs a documented decision rule rather than post-hoc tuning.

  • Report decided-case metrics together with abstention and full-policy recall
  • Keep Development evaluation separate from independent confirmation
  • Describe perturbation simulations as simulations, not observed rescans
  • Treat stable coefficient direction as benchmark association, never causation

Plain answers

Questions about score uncertainty

Does an indeterminate result mean the scan failed?

No. Technical capture can succeed while the measured orientation remains too close to the predefined decision boundary for a responsible binary label.

Did the uncertainty band make the model more accurate?

It changed the decision policy. Metrics among decided cases improved because ambiguous cases were withheld, while coverage fell and unresolved cases still counted against full-policy recall.

Do the perturbation results prove that live scores never change?

No. They cover deterministic ±5% count jitters in a frozen simulation. They do not model browser changes, website edits, network behavior or a fresh capture.

Source notes

References used for this guide

We prefer first-party standards, primary documentation and a visible interpretation boundary. Links are provided for verification and deeper implementation work.

Public uncertainty evaluation aggregate (JSON)

A domain-free public extract containing repeated metrics, indeterminate analysis, perturbation results and the frozen source-artifact hash.

Visible-feature evaluation script

The deterministic evaluation implementation used to generate the published aggregate artifact.

Public technical-yield aggregate (JSON)

A domain-free public extract containing all counts, outcome shares, cohort yields, intervals and the frozen source-artifact hash used in this brief.

VibeFootprint methodology

Defines the public-surface evidence boundary and the responsible interpretation of a Vibe-Footprint.

Apply the framework

Review a real public website.

See its pattern-similarity index, evidence breadth, separate security baseline and concrete findings.

Buy launch scan · €4.99