Files
bxh/docs/kg-redesign/implementation_plan.md

83 lines
1.7 KiB
Markdown
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

# new2 改造执行计划
## 阶段 0干净复刻
已完成:
-`原 new 目录` 复制到 `项目根目录`
- 排除 `node_modules``.env``data``__pycache__`、运行产物。
- 将空间汇报材料和基准测试材料纳入 `docs/reports`
## 阶段 1数据层补强
优先新增,不破坏旧表:
- `kg_entities`
- `kg_events`
- `kg_concepts`
- `kg_relations`
- `kg_statements`
- `kg_evidence_links`
- `kg_schema_proposals`
- `kg_place_spatial`
- `kg_geo_cells`
- `kg_route_metrics`
旧表兼容:
- `candidate_entities` 继续保留,用于旧流程。
- `candidate_relations` 可逐步迁移到 `kg_statements`
- `social_evidence` 保留,后续作为 Evidence 输入源。
- `ExperienceTag` 后续迁移为 `Concept`
## 阶段 2统一抽取 Agent
新增 `unified_kg_extract`,输入多源 evidence输出
```text
Entity / Event / Concept / Relation / Statement / SchemaProposal
```
暂不删除旧 `event_miner``xhs_agent``douyin_agent`,先让它们把原始证据写入 Evidence再由统一抽取层处理。
## 阶段 3Schema Auto Proposal
实现:
```text
schema_proposals
-> proposal merge
-> evidence count
-> reviewer approve/reject
-> schema version publish
```
禁止模型直接改生产 schema。
## 阶段 4空间能力
新增:
- PostGIS extension。
- `geom geometry(Point, 4326)`
- `h3_r7/h3_r8/h3_r9/h3_r10`
- `route_metrics` 缓存。
查询策略:
```text
H3 recall -> PostGIS ST_DWithin -> route topK -> KG semantic rank
```
## 阶段 5前端工作台
`/admin/plaza/overview` 中不需要的展示型功能逐步替换为:
- 数据源管理
- Evidence 浏览
- 候选知识审核
- Schema Proposal 审核
- 空间推荐测试台
- 图谱浏览