Train Neo4j Knowledge Graph From Website
lyzr_kg_ext_train_website_neo4jCrawl website URLs and ingest content into a Neo4j knowledge graph to train a RAG system, enabling knowledge extraction from web sources.
Instructions
Crawl and ingest websites into the Neo4j knowledge graph (namespaced endpoint).
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| urls | Yes | URLs to crawl and ingest | |
| actor | No | Apify actor to use (default "apify/website-content-crawler") | |
| rag_id | Yes | The ID of the RAG system to train | |
| source | No | Source label for the ingested content (default "website") | |
| chunk_size | No | Chunk size for splitting content (default 1000) | |
| crawler_type | No | Crawler type (default "cheerio") | |
| extra_fields | No | Any additional fields merged into the request body | |
| chunk_overlap | No | Chunk overlap for splitting content (default 100) | |
| max_crawl_depth | No | Maximum crawl depth (default 0) | |
| max_crawl_pages | No | Maximum number of pages to crawl (default 1) | |
| dynamic_content_wait_secs | No | Seconds to wait for dynamic content (default 5) |