Mapping Argentina’s Scientific Ecosystem: Building Collaboration Networks with Neo4j
summary
Uses open ministry data from 2011–2019 to build a temporal, tripartite knowledge graph of researchers–projects–subareas, integrating >45,000 researchers, ~15,000 projects and 7.5M+ collaboration links where collaboration = shared project. ETL disambiguates names by comparing local network neighborhoods. Database is Neo4j, natural language layer uses LangChain with local models and Gemini Pro/Flash to generate Cypher and summaries. Prototype UI supports Spanish queries and plans to add embeddings for semantic search.
AI-written summary · about
chapters (7)
- 0:00Introduction and motivation
- 2:09Data sources and scope
- 4:39Choosing Neo4j for analysis
- 6:49Building the data model
- 9:18Analyzing collaboration networks
- 10:47Natural language interface demo
- 15:01Future plans and next steps
related talks
15mAgentic GraphRAG: Simplifying Retrieval Across Structured & Unstructured Data
39mEnhancing RAG with Neo4j Knowledge Graph
29mSo... Graph Databases
28mSearching for Research Fraud in OpenAlex with Graph Data Science
41mManaging Connected Big Data in Art with Neo4j Graph Database
28mInterview with
49mWarming up to Graphs
14mDatashare + Neo4j: Unstructured Docs to Knowledge Graphs for Investigative Journalism