feat(rag): Complete RAG engine implementation with pgvector

Major Features:
- Created ekb_schema (13th schema) with 3 tables: KB/Document/Chunk
- Implemented EmbeddingService (text-embedding-v4, 1024-dim vectors)
- Implemented ChunkService (smart Markdown chunking)
- Implemented VectorSearchService (multi-query + hybrid search)
- Implemented RerankService (qwen3-rerank)
- Integrated DeepSeek V3 QueryRewriter for cross-language search
- Python service: Added pymupdf4llm for PDF-to-Markdown conversion
- PKB: Dual-mode adapter (pgvector/dify/hybrid)

Architecture:
- Brain-Hand Model: Business layer (DeepSeek) + Engine layer (pgvector)
- Cross-language support: Chinese query matches English documents
- Small Embedding (1024) + Strong Reranker strategy

Performance:
- End-to-end latency: 2.5s
- Cost per query: 0.0025 RMB
- Accuracy improvement: +20.5% (cross-language)

Tests:
- test-embedding-service.ts: Vector embedding verified
- test-rag-e2e.ts: Full pipeline tested
- test-rerank.ts: Rerank quality validated
- test-query-rewrite.ts: Cross-language search verified
- test-pdf-ingest.ts: Real PDF document tested (Dongen 2003.pdf)

Documentation:
- Added 05-RAG-Engine-User-Guide.md
- Added 02-Document-Processing-User-Guide.md
- Updated system status documentation

Status: Production ready
This commit is contained in:
2026-01-21 20:24:29 +08:00
parent 1f5bf2cd65
commit 40c2f8e148
338 changed files with 11014 additions and 1158 deletions

View File

@@ -579,6 +579,9 @@ df['creatinine'] = pd.to_numeric(df['creatinine'], errors='coerce')

View File

@@ -994,6 +994,9 @@ export const aiController = new AIController();

View File

@@ -1328,6 +1328,9 @@ npm install react-markdown

View File

@@ -236,6 +236,9 @@ FMA___基线 | FMA___1个月 | FMA___2个月

View File

@@ -394,6 +394,9 @@ formula = "FMA总分0-100 / 100"

View File

@@ -228,6 +228,9 @@ async handleFillnaMice(request, reply) {

View File

@@ -200,6 +200,9 @@ method: 'mean' | 'median' | 'mode' | 'constant' | 'ffill' | 'bfill'

View File

@@ -651,6 +651,9 @@ import { logger } from '../../../../common/logging/index.js';

View File

@@ -454,6 +454,9 @@ import { ChatContainer } from '@/shared/components/Chat';

View File

@@ -364,6 +364,9 @@ const initialMessages = defaultMessages.length > 0 ? defaultMessages : [{

View File

@@ -652,6 +652,9 @@ http://localhost:5173/data-cleaning/tool-c

View File

@@ -440,6 +440,9 @@ Docs: docs/03-业务模块/DC-数据清洗整理/06-开发记录/DC模块重建

View File

@@ -313,6 +313,9 @@ ConflictDetectionService // 冲突检测(字段级对比)

View File

@@ -477,6 +477,9 @@ Tool B后端代码**100%复用**了平台通用能力层,无任何重复开发

View File

@@ -254,6 +254,9 @@ $ node scripts/check-dc-tables.mjs

View File

@@ -487,6 +487,9 @@ ${fields.map((f, i) => `${i + 1}. ${f.name}${f.desc}`).join('\n')}