<i>Trans</i> determination of edible oils by fourier transform near‐infrared spectroscopy
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Abstract A generalized partial‐least‐squares calibration for determination of the trans content of edible fats and oils by Fourier transform near‐infrared (FT‐NIR) spectroscopy using 8‐mm disposable glass vials for sample handling and measurement was developed. The trans contents of a broad range of oils were determined using the American Oil Chemists' Society single‐bounce horizontal attenuated total reflectance (SB‐HATR) mid‐infrared spectroscopic procedure, these trans reference data were used in the development of the generalized FT‐NIR calibration. Additional refined and product‐specific calibrations were also developed, and all the calibrations were assessed for their predictive capabilities using two sets of validation samples, one comprising a broad range of oil types and the other restricted to oils with specific characteristics. The FT‐NIR trans predictions obtained using the generalized calibration were in good agreement with the SB‐HATR results; the values were accurate and reproducible to within ±1.1 and ±0.5% trans , respectively, compared to a reproducibility of ±0.40% trans obtained for the SB‐HATR method. The accuracy of the predictions obtained from the generalized FT‐NIR calibration for particular oil types was not significantly improved by supplementing the base training set with samples of these specific types. Calibrating only these oil types did, however, produce a substantial improvement in predictive accuracy, aproaching that of the SB‐HATR method. These product specific calibrations produced serious predictive errors when nonrepresentative samples were analyzed. The incorporation of a supplementary discriminate analysis routine was found to be a powerful safeguard in flagging nonrepresentative samples as outliers and could also be used to select the calibration most appropriate for the characteristics of the sample being analyzed. Overall, it was concluded that FT‐NIR spectroscopy provides a viable alternative to the SB‐HATR/mid‐Fourier transform infrared method for trans determination, making use of more industrially robust instrumentation and equipped with a simpler sample handling system.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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 it