/**
 * Metric Type Definitions
 *
 * All metrics collected by experiments must conform to these types.
 * Each metric type corresponds to a LaTeX table in the thesis.
 */
/**
 * Statistical test result (Mann-Whitney U, Cohen's d, etc.)
 */
export interface StatisticalTestMetric {
    comparison: string;
    method1Mean: number;
    method2Mean: number;
    statistic: string;
    u: number;
    pValue: number;
    cohensD: number;
}
/**
 * Hub traversal efficiency metric
 */
export interface HubTraversalMetric {
    dataset: string;
    nodes: number;
    method: string;
    hubTraversal: number;
    paths?: number;
}
/**
 * Path length distribution metric
 */
export interface PathLengthMetric {
    dataset: string;
    method: string;
    min: number;
    max: number;
    mean: number;
    median: number;
}
/**
 * Runtime performance metric
 */
export interface RuntimePerformanceMetric {
    dataset: string;
    nodes: number;
    dpTime: number;
    bfsTime: number;
    dpNodesPerSec?: number;
    bfsNodesPerSec?: number;
}
/**
 * Scalability metric across graph sizes
 */
export interface ScalabilityMetric {
    dataset: string;
    nodes: number;
    dpTime: number;
    bfsTime: number;
    ratio: number;
}
/**
 * Cross-dataset path diversity comparison
 */
export interface CrossDatasetMetric {
    dataset: string;
    nodes: number;
    dpDiversity: number;
    bfsDiversity: number;
    improvement: number;
}
/**
 * Method ranking by path diversity
 */
export interface MethodRankingMetric {
    method: string;
    diversity: number;
    paths: number;
}
/**
 * Perturbation consistency metric
 */
export interface PerturbationMetric {
    perturbation: string;
    dpDiversity: number;
    bfsDiversity: number;
    winner: string;
}
/**
 * Structural representativeness metric (ego network)
 */
export interface StructuralRepresentativenessMetric {
    coverage: number;
    precision: number;
    f1Score: number;
    intersectionSize: number;
    totalNodes: number;
}
/**
 * Structural representativeness metrics (hub coverage)
 */
export interface StructuralRepresentativenessMetricsMetric {
    totalSampled: number;
    hubCoverage: number;
    bucketsCovered: number;
    totalBuckets: number;
}
/**
 * N-seed generalization metric
 */
export interface NSeedGeneralizationMetric {
    n: number;
    variant: string;
    nodes: number;
    paths: number;
}
/**
 * N-seed comparison across methods
 */
export interface NSeedComparisonMetric {
    method: string;
    n: number;
    nodes: number;
    paths: number;
    iterations: number;
    coverage: number;
}
/**
 * N-seed hub traversal comparison
 */
export interface NSeedHubTraversalMetric {
    graph: string;
    method: string;
    paths: number;
    hubTraversal: number;
}
/**
 * N-seed path diversity comparison
 */
export interface NSeedPathDiversityMetric {
    graph: string;
    method: string;
    paths: number;
    uniqueNodes: number;
    diversity: number;
}
/**
 * Algorithm comparison metric
 */
export interface AlgorithmComparisonMetric {
    graph: string;
    n: number;
    method: string;
    nodes: number;
    paths: number;
    diversity: number;
}
/**
 * Hub traversal comparison metric
 */
export interface HubTraversalComparisonMetric {
    graph: string;
    method: string;
    hubTraversal: number;
}
/**
 * MI ranking quality metric (Path Salience Ranking)
 */
export interface MIRankingQualityMetric {
    dataset: string;
    meanMI: number;
    nodeCoverage: number;
    pathDiversity: number;
    paths: number;
}
/**
 * Ranking benchmarks metric
 */
export interface RankingBenchmarksMetric {
    dataset: string;
    method: string;
    meanMI: number;
    nodeCoverage: number;
    pathDiversity: number;
}
/**
 * Hub mitigation analysis metric
 */
export interface HubMitigationMetric {
    method: string;
    nodes: number;
    paths: number;
    iterations: number;
}
/**
 * Multi-hub expansion efficiency metric
 */
export interface MultiHubEfficiencyMetric {
    method: string;
    nodesExpanded: number;
    hubsExpanded: number;
    hubRatio: number;
    pathsFound: number;
}
/**
 * Hub-avoidance metrics for evaluating degree-prioritised expansion.
 *
 * Measures how effectively an algorithm avoids expanding through high-degree
 * hub nodes during traversal. Complements path diversity by directly measuring
 * the design goal of hub avoidance.
 */
export interface HubAvoidanceMetric {
    /** Dataset identifier */
    dataset: string;
    /** Algorithm method name */
    method: string;
    /** Number of seeds (N) */
    n: number;
    /** Proportion of expanded nodes that are hubs (0-1, lower is better) */
    hubTraversalRate: number;
    /** Ratio of peripheral nodes expanded to hub nodes expanded (higher is better) */
    peripheralCoverageRatio: number;
    /** Total nodes expanded */
    totalExpanded: number;
    /** Number of hub nodes expanded (degree >= threshold) */
    hubCount: number;
    /** Number of peripheral nodes expanded (degree <= threshold) */
    peripheralCount: number;
    /** Optional: degree distribution as JSON string */
    degreeDistribution?: string;
}
/**
 * Salience coverage comparison metric
 *
 * Measures how well expansion algorithms discover high-salience paths
 * (as ranked by Path Salience algorithm using mutual information).
 */
export interface SalienceCoverageComparisonMetric {
    /** Dataset identifier */
    dataset: string;
    /** Algorithm method name */
    method: string;
    /** Number of seeds (N) */
    n: number;
    /** Salience coverage: fraction of top-K paths found (0-1) */
    salienceCoverage: number;
    /** Salience precision: of discovered paths, fraction that are in top-K (0-1) */
    saliencePrecision: number;
    /** Number of top-K salient paths found */
    topKFound: number;
    /** Total number of top-K salient paths (ground truth) */
    topKTotal: number;
    /** Total paths discovered by the algorithm */
    pathsDiscovered: number;
    /** Total nodes expanded during traversal */
    nodesExpanded: number;
    /** Number of expansion iterations */
    iterations: number;
    /** Runtime in milliseconds */
    runtimeMs: number;
}
/**
 * Classification correctness metric (per-class precision/recall/F1)
 */
export interface ClassificationCorrectnessMetric {
    graphClass: string;
    precision: number;
    recall: number;
    f1: number;
    support: number;
}
/**
 * Classification significance metric (overall accuracy vs baseline)
 */
export interface ClassificationSignificanceMetric {
    application: string;
    baselineAccuracy: number;
    classifierAccuracy: number;
}
/**
 * Generation correctness metric (per-class acceptance rate)
 */
export interface GenerationCorrectnessMetric {
    graphClass: string;
    accepted: number;
    total: number;
    acceptanceRate: number;
    meanConfidence: number;
}
/**
 * Generation significance metric (overall rate vs baseline)
 */
export interface GenerationSignificanceMetric {
    metric: string;
    achieved: number;
    randomBaseline: number | null;
}
/**
 * Ranking correctness metric (per-method MI quality)
 */
export interface RankingCorrectnessMetric {
    method: string;
    meanMI: number;
    nodeCoverage: number;
    pathDiversity: number;
    spearmanRho: number;
}
/**
 * Ranking significance metric (pairwise baseline comparisons)
 */
export interface RankingSignificanceMetric {
    baseline: string;
    miImprovement: number;
    wins: string;
    meanMI: number;
}
/**
 * Community detection metric (Louvain, Leiden, Label Propagation)
 */
export interface CommunityDetectionMetric {
    dataset: string;
    method: string;
    communities: number;
    modularity: number;
    iterations: number;
    nodes: number;
}
/**
 * K-core decomposition metric
 */
export interface KCoreDecompositionMetric {
    dataset: string;
    degeneracy: number;
    coreCount: number;
    maxCoreSize: number;
    nodes: number;
    edges: number;
}
/**
 * Union of all metric types for internal handling
 */
export type Metric = StatisticalTestMetric | HubTraversalMetric | PathLengthMetric | RuntimePerformanceMetric | ScalabilityMetric | CrossDatasetMetric | MethodRankingMetric | PerturbationMetric | StructuralRepresentativenessMetric | StructuralRepresentativenessMetricsMetric | NSeedGeneralizationMetric | NSeedComparisonMetric | NSeedHubTraversalMetric | NSeedPathDiversityMetric | AlgorithmComparisonMetric | HubTraversalComparisonMetric | MIRankingQualityMetric | RankingBenchmarksMetric | HubMitigationMetric | MultiHubEfficiencyMetric | HubAvoidanceMetric | SalienceCoverageComparisonMetric | ClassificationCorrectnessMetric | ClassificationSignificanceMetric | GenerationCorrectnessMetric | GenerationSignificanceMetric | RankingCorrectnessMetric | RankingSignificanceMetric | CommunityDetectionMetric | KCoreDecompositionMetric;
/**
 * Metric category - groups related metrics for table generation
 */
export type MetricCategory = "statistical-significance" | "hub-traversal" | "hub-avoidance" | "path-lengths" | "runtime-performance" | "scalability" | "cross-dataset" | "method-ranking" | "perturbation" | "structural-representativeness" | "structural-representativeness-metrics" | "n-seed-generalization" | "n-seed-comparison" | "n-seed-hub-traversal" | "n-seed-path-diversity" | "algorithm-comparison" | "hub-traversal-comparison" | "mi-ranking-quality" | "ranking-benchmarks" | "hub-mitigation" | "multi-hub-efficiency" | "salience-coverage-comparison" | "classification-correctness" | "classification-significance" | "generation-correctness" | "generation-significance" | "ranking-correctness" | "ranking-significance" | "community-detection" | "k-core-decomposition";
/**
 * Typed metric record with category
 */
export interface MetricRecord<T = Metric> {
    category: MetricCategory;
    data: T;
    timestamp?: string;
}
/**
 * Complete metrics output structure
 */
export interface MetricsOutput {
    version: string;
    timestamp: string;
    metrics: Partial<Record<MetricCategory, Metric[]>>;
}
//# sourceMappingURL=types.d.ts.map