SCOPE and SCION
What it is
A framework for schema induction and ontology fusion from text. It includes SCOPE, a benchmark for converting relation and event extraction datasets into a unified schema-edge representation, and SCION, an auditable pipeline for generating and evaluating schema graphs.
The problem it solves
The difficulty of extracting consistent, high-quality schemas and ontologies from raw text and disparate datasets.
What you could use it for
- Converting public relation extraction (RE) and event extraction (EE) datasets into unified typed schemas.
- Inducing and fusing schemas from text with contract constraints.
- Evaluating schema graphs using multiple metrics (Literal, Fuzzy, Continuous, and Graph).
- Generating knowledge graphs from text using the SCION pipeline.
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.