Synthetic vs Natural Indicator Matrix Scorer
Combine caller independently calibrated indicator likelihoods/prior under explicit conditional-independence model, retaining absent evidence; correlated observations invalidate naïve combination and result is not natural/synthetic certification.
Description
Bayesian evidence combines a prior with likelihoods under an explicit conditional-independence model.
Bayesian evidence combines a prior with likelihoods under an explicit conditional-independence model. An absent indicator contributes its complementary likelihood and must not be discarded.1
Combine caller independently calibrated indicator likelihoods/prior under explicit conditional-independence model, retaining absent evidence; correlated observations invalidate naïve combination and result is not natural/synthetic certification.
When to use Synthetic vs Natural Indicator Matrix Scorer
Use this tool to combine caller independently calibrated indicator likelihoods/prior under explicit conditional-independence model, retaining absent evidence; correlated observations invalidate naïve combination and result is not natural/synthetic certification.
Prepare the gemstone measurements, spectra or comparison data and any supplied calibration parameters. Check units, reference conditions and record alignment.
How to use Synthetic vs Natural Indicator Matrix Scorer and understand its results
- Prepare the gemstone measurements, spectra or comparison data and any supplied calibration parameters. Check units, reference conditions and record alignment.
- Enter priorSyntheticProbability, observedIndicatorFlags, syntheticIndicatorProbabilities, naturalIndicatorProbabilities using the field table and calculation rules, then run the tool.
- Read independentModelSyntheticProbability, logEvidenceLikelihoodRatio. The explanation below defines the units, signs, ordering and any flags or placeholder values.
- If the tool reports an input issue, correct the values and run it again. Apply the stated assumptions when interpreting a successfully returned result.
| Input field | Meaning | Value type; see rules below | Requirement |
|---|---|---|---|
| priorSyntheticProbability | Prior synthetic probability | number | Required |
| observedIndicatorFlags | Observed indicator flags | list of number | Required |
| syntheticIndicatorProbabilities | Synthetic indicator probabilities | list of number | Required |
| naturalIndicatorProbabilities | Natural indicator probabilities | list of number | Required |
Pair binary observations with synthetic/natural indicator probabilities, all strictly between 0 and 1, and a prior in the same range. Accumulate present and absent log-likelihood ratios and use a stable logistic conversion. Return model probability and evidence log ratio.
Paired lists must be nonempty and have equal lengths.
posterior log odds=prior log odds+Σlog[P(observed flag|synthetic)/P(observed flag|natural)] Units, indexing, boundary conditions and exceptional cases are defined in the calculation rules above.
| Output field | Meaning | Value type |
|---|---|---|
| independentModelSyntheticProbability | Independent model synthetic probability | number |
| logEvidenceLikelihoodRatio | Log evidence likelihood ratio | number |
{
"input": {
"priorSyntheticProbability": 0.5,
"observedIndicatorFlags": [
1,
0
],
"syntheticIndicatorProbabilities": [
0.8,
0.5
],
"naturalIndicatorProbabilities": [
0.2,
0.5
]
},
"output": {
"independentModelSyntheticProbability": 0.8,
"logEvidenceLikelihoodRatio": 1.3862943611198906
}
}Limitations and assumptions
Likelihoods must already be independently calibrated. Correlated observations can invalidate the combination; it is not origin certification.
Numeric inputs and returned numeric values must be finite; an unrepresentable returned value produces an error. Numeric lists are limited to 100,000 entries; the rules above may require smaller limits. Missing required fields, extra fields, wrong value types and violations of the calculation rules are input errors.
Alternative or Complementary methods
Use UV Reactivity Note Generator: Produce bounded observation-only UV reactivity note from measured post-excitation frames and caller detection policy; no lifetime extrapolation, phosphorescence mechanism, identity or UV handling advice. For a complementary task, use Fluorescence Strength Grade Mapper: Map excitation-normalized background-corrected fluorescence under caller assay-specific bands; retain negative nondetections and no commercial UV grade/material identity or safe lamp exposure inferred.
References
Similar or alternative tools
- UV Reactivity Note Generator
Produce bounded observation-only UV reactivity note from measured post-excitation frames and caller detection policy; no lifetime extrapolation, phosphorescence mechanism, identity or UV handling advice.
- Fluorescence Strength Grade Mapper
Map excitation-normalized background-corrected fluorescence under caller assay-specific bands; retain negative nondetections and no commercial UV grade/material identity or safe lamp exposure inferred.