Human Schema Curation via Causal Association Rule Mining

TL;DR

This paper introduces a human-in-the-loop framework for constructing event schema libraries using causal association rule mining, enabling non-experts to create detailed scenario models efficiently.

Contribution

It presents a novel script induction system and interface for non-experts to program complex event structures, along with a publicly available schema library of 232 detailed schemas.

Findings

Created a schema library with 232 detailed event schemas

Developed a user-friendly interface for schema curation

Demonstrated effective human-in-the-loop schema construction

Abstract

Event schemas are structured knowledge sources defining typical real-world scenarios (e.g., going to an airport). We present a framework for efficient human-in-the-loop construction of a schema library, based on a novel script induction system and a well-crafted interface that allows non-experts to "program" complex event structures. Associated with this work we release a schema library: a machine readable resource of 232 detailed event schemas, each of which describe a distinct typical scenario in terms of its relevant sub-event structure (what happens in the scenario), participants (who plays a role in the scenario), fine-grained typing of each participant, and the implied relational constraints between them. We make our schema library and the SchemaBlocks interface available online.