@codai/cbd/dist/features/advanced-vector-search.d.ts

Codai Better Database - High-Performance Vector Memory System with HPKV-inspired architecture and MCP server

/**
 * Advanced Vector Search Engine
 * Enhanced capabilities for hybrid search, multi-modal embeddings, and optimization
 */
import { EventEmitter } from 'events';
interface HybridSearchOptions {
    vectorWeight?: number;
    keywordWeight?: number;
    semanticWeight?: number;
    maxResults?: number;
    minSimilarity?: number;
    includeMetadata?: boolean;
}
interface MultiModalEmbedding {
    text?: number[];
    image?: number[];
    audio?: number[];
}
interface VectorCluster {
    id: string;
    centroid: number[];
    members: string[];
    similarity: number;
}
declare class AdvancedVectorSearchEngine extends EventEmitter {
    private openai;
    private vectorIndex;
    private metadataIndex;
    private keywordIndex;
    private clusters;
    private searchCache;
    constructor(config: {
        openaiApiKey: string;
        cacheSize?: number;
        clusterThreshold?: number;
    });
    private initializeAdvancedFeatures;
    /**
     * Hybrid Search: Combine vector, keyword, and semantic search
     */
    hybridSearch(query: string, options?: HybridSearchOptions): Promise<{
        results: Array<{
            id: string;
            content: string;
            score: number;
            matchType: 'vector' | 'keyword' | 'semantic' | 'hybrid';
            metadata?: any;
        }>;
        performance: {
            totalTime: number;
            vectorTime: number;
            keywordTime: number;
            semanticTime: number;
        };
    }>;
    /**
     * Multi-modal Vector Embeddings
     */
    generateMultiModalEmbedding(content: {
        text?: string;
        imageUrl?: string;
        audioUrl?: string;
    }): Promise<MultiModalEmbedding>;
    /**
     * Dynamic Vector Index Optimization
     */
    optimizeVectorIndex(): Promise<{
        clustersCreated: number;
        indexReorganized: boolean;
        performanceImprovement: number;
    }>;
    /**
     * Batch Vector Operations
     */
    batchVectorOperations(operations: Array<{
        type: 'insert' | 'update' | 'delete' | 'search';
        id?: string;
        vector?: number[];
        query?: string;
        metadata?: any;
    }>): Promise<{
        results: any[];
        performance: {
            totalTime: number;
            operationsPerSecond: number;
        };
    }>;
    /**
     * Vector Similarity Clustering
     */
    performVectorClustering(options?: {
        minClusterSize?: number;
        maxClusters?: number;
        similarityThreshold?: number;
    }): Promise<VectorCluster[]>;
    private performVectorSearch;
    private performKeywordSearch;
    private performSemanticSearch;
    private combineSearchResults;
    private generateCacheKey;
    private setupDynamicClustering;
    private setupSearchCache;
    private setupPerformanceMonitoring;
    private generateImageEmbedding;
    private generateAudioEmbedding;
    private analyzeIndexStructure;
    private createVectorClusters;
    private reorganizeIndex;
    private measurePerformanceImprovement;
    private groupOperationsByType;
    private executeBatchOperation;
    private createCluster;
}
export { AdvancedVectorSearchEngine, HybridSearchOptions, MultiModalEmbedding, VectorCluster };
//# sourceMappingURL=advanced-vector-search.d.ts.map