# Beef electronic-nose benchmark found lower scores when whole samples were held out

> A Foods reanalysis of 12 beef cuts separated complete sensor acquisitions during testing. Feature summaries explained much of the difference between tree and neural models.

URL: https://foodradar.org/leakage-free-benchmarking-reveals-realistic-accuracy-limits-for-electronic-nose-beef-freshness-models
Language: en
Author: FoodRadar Editorial
Published: 2026-10-19T06:00:10.000Z
Updated: 2026-10-25T01:38:37.766Z
Source publication date: 2026-08-10
Category: Technology
Türkçe: https://foodradar.org/tr/leakage-free-benchmarking-reveals-realistic-accuracy-limits-for-electronic-nose-beef-freshness-models
Español: https://foodradar.org/es/leakage-free-benchmarking-reveals-realistic-accuracy-limits-for-electronic-nose-beef-freshness-models
Deutsch: https://foodradar.org/de/leakage-free-benchmarking-reveals-realistic-accuracy-limits-for-electronic-nose-beef-freshness-models
Italiano: https://foodradar.org/it/leakage-free-benchmarking-reveals-realistic-accuracy-limits-for-electronic-nose-beef-freshness-models
Русский: https://foodradar.org/ru/leakage-free-benchmarking-reveals-realistic-accuracy-limits-for-electronic-nose-beef-freshness-models

## Twelve cuts, tested separately

A study by Erkan Caner Ozkat, published in Foods on 10 August 2026, reanalysed public electronic-nose data for beef freshness. The primary dataset contains 12 cuts recorded with 11 metal-oxide gas-sensor channels. Each test held out one complete cut while models trained on the others; total viable count, or TVC, defined the freshness labels but was excluded from model inputs.

Across these tests, the gradient-boosted-tree pipeline achieved accuracy of 0.81 ± 0.10 and macro-F1 of 0.68 ± 0.15. The raw-sequence neural network reached macro-F1 of 0.50 ± 0.15. When the same network was evaluated using random splits instead, macro-F1 rose to 0.97. A separate comparison supplying TVC as an input produced near-perfect random-forest accuracy; the author does not establish that these procedures caused any specific earlier publication’s result.

## Features and fewer sensor channels

When a neural network received the same window summaries as the tree, it reached macro-F1 of 0.64 ± 0.17, with no statistically significant difference from the tree in the paired comparison. A five-channel subset selected within training folds reached macro-F1 of 0.64, versus 0.68 for the full array; that difference was also not significant. The author states that a physical reduced array still needs construction and validation.

Separate coffee and aroma datasets tested the role of richer heater-profile signals. On the coffee task, a position-aware neural network exceeded the tree, and its advantage reversed when the profile was reduced to one step. On the seven-class aroma task, the tree remained stronger. The author describes these datasets as signal-structure comparisons rather than additional meat-quality validation.

## Reference data and transfer limits

Ten of the 12 cuts share a partially identical TVC reference series. The author therefore distinguishes validation across 12 sensor acquisitions from validation across 12 independent microbial trajectories. On a separate beef acquisition, transfer macro-F1 was 0.40 for the tree and 0.46 for the neural network. The author reports that per-device standardisation improved the network, using unlabelled recordings from the target acquisition.

The author calls for broader validation across laboratories, storage conditions and longer periods, noting that the 37-hour acquisitions cannot characterise long-term sensor drift. The paper describes TVC as a freshness measure rather than a measure of pathogen or toxin presence. The study was supported by the Recep Tayyip Erdoğan University Development Foundation, grant 02026007008461; the author declared no conflicts of interest.

## Source and licence

Source: Erkan Caner Ozkat. Leakage-Free Benchmarking of Electronic Noses for Beef Freshness: A Signal-Richness Criterion for Model Selection. Foods (2026), 15, 2798. DOI: 10.3390/foods15162798. © 2026 by the author. CC BY 4.0. FoodRadar prepared and edited this summary.

## Sources and rights

- [Leakage-Free Benchmarking of Electronic Noses for Beef Freshness: A Signal-Richness Criterion for Model Selection](https://doi.org/10.3390/foods15162798?utm_source=foodradar.org&utm_medium=referral&utm_campaign=editorial&utm_content=leakage-free-benchmarking-reveals-realistic-accuracy-limits-for-electronic-nose-beef-freshness-models-en). Foods. Erkan Caner Ozkat (2026). Leakage-Free Benchmarking of Electronic Noses for Beef Freshness: A Signal-Richness Criterion for Model Selection. Foods 15(16), 2798. DOI: 10.3390/foods15162798. © 2026 by the author. CC BY 4.0. Adapted, edited and translated by FoodRadar. [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/?utm_source=foodradar.org&utm_medium=referral&utm_campaign=editorial&utm_content=leakage-free-benchmarking-reveals-realistic-accuracy-limits-for-electronic-nose-beef-freshness-models-en-license).

Adapted by FoodRadar.

[Editorial standards](https://foodradar.org/editorial-policy)