Labor Market Landscape Modeling with Knowledge Graphs
summary
A production Job‑Skills Knowledge Graph (JSKG) built on Neo4j ingests the national Skills Framework, near‑real‑time job postings and curated training courses, uses in‑house ML to extract jobs/skills, human labels sectors, and attaches computed similarity weights to power career‑transition recommendations; a graph neural network link‑prediction step uncovers implicit task–skill links; nontechnical users explore results via Neo4j Bloom and the JSKG backs the Singapore job skills portal; code not yet open source.
AI-written summary · about
chapters (13)
- 0:00Introduction and speakers
- 0:24About the presenters
- 1:29Motivation for labor market modeling
- 2:50Job Skills Knowledge Graph overview
- 3:54JSKG entities and relationships
- 5:17Value for analysts and users
- 6:23Career transition pathways demo
- 6:45Graph relational foundation models
- 8:30Transforming relational data to graphs
- 9:34Predictive insights with graph models
- 10:38Contributions and publications
- 11:20Challenges and future directions
- 12:03Q&A session
related talks
01h 19mIntro to GraphRAG
44mBuilding Knowledge Graphs
24mWhy You Need a Knowledge Graph And How to Build It
25mKnowledge graphs for financial institutions
46mComputing Professional Identity for the Economic Graph
46mBuilding Knowledge Graphs- 44mBuilding Knowledge Graphs
21mMastering Knowledge Graph Transformations Without Compromising Meaning