import type { ComposedWord } from "./ComposedWord.js";
import type { SingleWord } from "./SingleWord.js";
/**
 * Splits raw text into sentence strings.
 */
export interface SentenceSplitter {
    split(text: string): string[];
}
/**
 * Splits a sentence or text fragment into tokens.
 */
export interface Tokenizer {
    tokenize(text: string): string[];
}
/**
 * Loads a stopword set for a language.
 */
export interface StopwordProvider {
    load(language: string): Set<string>;
}
/**
 * Compares two candidate strings for deduplication.
 */
export interface SimilarityStrategy {
    compare(a: string, b: string): number;
}
/**
 * Context passed to candidate normalizers.
 */
export interface CandidateNormalizerContext {
    readonly original: string;
    readonly language: string;
}
/**
 * Normalizes tokens before candidate aggregation.
 */
export interface CandidateNormalizer {
    normalize(token: string, context: CandidateNormalizerContext): string;
}
/**
 * Context passed to lemmatizers.
 */
export interface LemmatizerContext {
    readonly original: string;
    readonly language: string;
}
/**
 * Optional token lemmatization hook.
 */
export interface Lemmatizer {
    lemmatize(token: string, context: LemmatizerContext): string;
}
/**
 * Candidate metadata available to user-provided filters.
 */
export interface CandidateFilterInput {
    readonly keyword: string;
    readonly normalizedKeyword: string;
    readonly ngramSize: number;
    readonly score: number;
    readonly occurrences: number;
    readonly sentenceIds: number[];
}
/**
 * Post-processing scorer hook for ranked candidates.
 */
export interface KeywordScorer {
    score(candidates: KeywordResult[]): KeywordResult[];
}
/**
 * Context passed to single-word scorers.
 */
export interface SingleWordScoreContext {
    readonly maxTf: number;
    readonly avgTf: number;
    readonly stdTf: number;
    readonly numberOfSentences: number;
    readonly features?: string[] | null;
}
/**
 * Hook for replacing the internal YAKE single-word score formula.
 */
export interface SingleWordScorer {
    score(term: SingleWord, context: SingleWordScoreContext): number;
}
/**
 * Context passed to multi-word scorers.
 */
export interface MultiWordScoreContext {
    readonly features?: string[] | null;
}
/**
 * Hook for replacing the internal YAKE multi-word score formula.
 */
export interface MultiWordScorer {
    score(candidate: ComposedWord, context: MultiWordScoreContext): number;
}
/**
 * Combined sentence-splitting and tokenization surface.
 */
export interface TextProcessor {
    splitSentences(text: string): string[];
    tokenizeWords(text: string): string[];
}
/**
 * Rich keyword record returned by detailed extraction APIs.
 */
export interface KeywordResult {
    keyword: string;
    normalizedKeyword: string;
    score: number;
    ngramSize: number;
    occurrences: number;
    sentenceIds: number[];
}
/**
 * Default sentence splitter used by Yaket.
 */
export declare const defaultSentenceSplitter: SentenceSplitter;
/**
 * Default tokenization strategy used by Yaket.
 */
export declare const defaultTokenizer: Tokenizer;
/**
 * Default stopword loader backed by bundled assets.
 */
export declare const defaultStopwordProvider: StopwordProvider;
/**
 * Default text processor combining the bundled sentence splitter and tokenizer.
 */
export declare const defaultTextProcessor: TextProcessor;
//# sourceMappingURL=strategies.d.ts.map