← All projects

PROJECT 035 / 102 · Food science

Quality model errors by wine type.

Does a pooled quality model have different error or bias for red and white samples?

Focused analytical studyWine qualityExecuted notebook

THE RESULT

What the data shows

The largest computed metric is 0.5465 for white; the smallest is 0.444 for red. Metric: mae (Sensory-score MAE).

This is a subgroup diagnostic of the shared holdout, not a separate independent validation.

Quality model errors by wine type — chart from the computed study output
Computed study output. Full values and units are available in the results download.
Computed results · 2 of 2 rows. Values rounded for display; source units and raw column names are retained in the download.
wine typetest samplesmaebias
red3340.4440.03391
white9540.5465-0.01428
Download the complete result table ↓

THE METHOD

From source to answer

Held-out residual summaries stratified by wine type for a fixed Extra Trees model.

The Python source exposes this study’s transformations. The complete project download includes shared preparation and evaluation routines.

EVALUATION

How the result was checked

Fixed 80/20 split of unique input groups (seed 42); exact input duplicates stay together.

5,209Training observations
1,288Test observations

Fixed configurations specified before scoring; no tuning on the holdout.

Target and input features

Target: quality

Inputs: fixed_acidity, volatile_acidity, citric_acid, residual_sugar, chlorides, free_sulfur_dioxide, total_sulfur_dioxide, density, ph, sulphates, alcohol, wine_type

Download evaluation record ↓

THE NEXT DECISION

What follows from the finding

Investigate subgroup calibration and obtain external batch-level validation before specialization.

Where the conclusion stops

Sensory scores are ordinal and concentrated in the middle. Producer and batch IDs are unavailable. Associations are not recipes for changing quality or evidence of market price.

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.