Leveraging machine learning for the optimization of reinforced rapeseed protein-gelatin edible coatings for enhanced food preservation
Bibliographic record
Abstract
• Expanded 13 data points to 1,000 using Gaussian noise for data augmentation. • ML models showed R 2 : 0.9517–0.9998 and MSE: 0.0001–0.5537. • Optimized film had T.S. 28.94 MPa, E% 22.61, and WVP 4.40 × 10 2 g·mm/m 2 ·day·kPa. • Model predictions showed strong correlation with experimental results. • Edible film coating extended fruit shelf life by 7 days at ambient conditions. Optimizing component ratios is crucial for enhancing bioplastic formulations. Traditional optimization is time-consuming, while data-driven methods require large datasets. To address this, we augmented thirteen experimental data points to 1,000, enhancing machine learning (ML) model accuracy. Three ML models were trained to optimize the formulation of a rapeseed protein-gelatin nanocomposite reinforced with cellulose nanocrystals (CNC) and cross-linked with citric acid. The models showed high predictive performance (R 2 : 0.9517–0.9998; MSE: 0.0001–0.5537; MAE: 0.0018–0.4221). The optimal formulation (1:1:5.5:3.87) yielded a film with 28.94 ± 1.18 MPa tensile strength, 22.61 ± 2.12 % elongation, and 4.40 ± 0.04 g mm/d·m 2 ·kPa × 10 2 water vapor permeability. A film based on this formulation was applied as a coating on fruits and showed an increased shelf life, verifying its practical application. Our results show that the integration of generated data enhances ML predictive performance, capturing complex nonlinear relationships between input and output variables, thus balancing the mechanical and barrier properties of biopolymer films. This work highlights the potential of data generation and ML in materials science by providing a pathway to accelerate the development of sustainable bio-based plastics.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".