Unified decision output for a single ClassifierQuestion.
Signature:
export declare interface ClassifierDecision
Properties
| Property | Type | Description |
|---|---|---|
| confidence | number | Normalized certainty confidence in [0.0, 1.0]. |
| expectedScore | number | Ordinal only: continuous expected score across levels, sum(level_i * p_i). |
| id | string | Question identifier matching ClassifierQuestion.id. |
| label | string | Winning option label ("true"/"false" for binary, winning label for categorical/ordinal). |
| predictionSet | string[] | Optional conformal prediction set covering >= 90% probability mass. |
| probabilities | ClassifierOptionProbability[] | Full calibrated probability distribution over candidate options. |
| probability | number | Binary only: calibrated P(true) in [0.0, 1.0]. |
ClassifierDecision.confidence
Normalized certainty confidence in [0.0, 1.0].
Signature:
confidence: number;
ClassifierDecision.expectedScore
Ordinal only: continuous expected score across levels, sum(level_i * p_i).
Signature:
expectedScore?: number;
ClassifierDecision.id
Question identifier matching ClassifierQuestion.id.
Signature:
id: string;
ClassifierDecision.label
Winning option label ("true"/"false" for binary, winning label for categorical/ordinal).
Signature:
label: string;
ClassifierDecision.predictionSet
Optional conformal prediction set covering >= 90% probability mass.
Signature:
predictionSet?: string[];
ClassifierDecision.probabilities
Full calibrated probability distribution over candidate options.
Signature:
probabilities: ClassifierOptionProbability[];
ClassifierDecision.probability
Binary only: calibrated P(true) in [0.0, 1.0].
Signature:
probability?: number;