Knowledge graph intelligence for the circular economy
Excess Materials Exchange was building a digital marketplace to give surplus materials a second life. The challenge went beyond connecting supply and demand: the platform needed to understand relationships between materials, properties, organisations and potential applications to identify meaningful reuse opportunities.
Circular Economy · ESG
2019–2021
Technology leadership · Knowledge graph · Recommender systems
Apr 2019 – Jun 2021
Approach
We engineered a graph-based intelligence layer that connected materials, organisations, properties and potential applications.
Rather than relying solely on conventional search, filtering or collaborative recommendations, the platform used relationships and context within the knowledge graph to discover relevant matches and reuse opportunities.
Engineered
Neo4j-based knowledge graph and semantic data model. Graph-powered recommendation engine using relational reasoning and knowledge graph embeddings. NLP pipelines for extracting and structuring material information. Hyperledger-based infrastructure supporting provenance and circular-economy tracking.
- Neo4j
- ·Knowledge Graphs
- ·Graph ML
- ·NLP
- ·Hyperledger
Outcomes
Smarter matching
Context-aware recommendations across materials and reuse opportunities.
Broader discovery
Potential matches beyond conventional search and filtering.
Explainable
Recommendations traceable through graph relationships.
Built for Excess Materials Exchange as part of its circular-economy technology platform