Add graphrag-neo4j module: structured-output entity and relation extraction into a typed Neo4j graph with a vector index, multi-hop retrieval by span linking and variable-length Cypher paths, compared with plain vector RAG over the same 35 chunks and 9 questions (complete evidence 7/9 vs 5/9 at k=8); Testcontainers on neo4j:2026.10.1 and 5.26. The extraction step is a rule-based stand-in, not a language model.

Co-Authored-By: Claude Sonnet 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_01CifR2BE7PBvRy5hT8TS9oz
2026-10-26 02:03:40 +00:00
parent 279be7eb9f
commit 108fc44542
66 changed files with 1997 additions and 0 deletions
@@ -0,0 +1,10 @@
+#!/usr/bin/env bash
+# Regenerates every file in output/ by running the test suite.
+# Needs: JDK 25, Maven, a running Docker daemon (Testcontainers starts Neo4j itself), internet for the first run
+# (about 90 MB for the MiniLM model, cached in ~/.cache/minilm, and the Neo4j image, about 400 MB).
+# NEO4J_IMAGE overrides the image, e.g. NEO4J_IMAGE=neo4j:5-community ./scripts/run-all.sh
+set -euo pipefail
+cd "$(dirname "$0")/.."
+rm -rf output
+mvn -B -q test
+ls output