ASER: A Large-scale Eventuality Knowledge Graph

TL;DR

ASER is a comprehensive large-scale knowledge graph capturing activities, states, and events from extensive textual data, filling a gap left by entity-focused graphs and supporting advanced language understanding.

Contribution

This work introduces ASER, a novel large-scale eventuality knowledge graph with diverse relations and extensive data, enhancing language understanding beyond entity knowledge.

Findings

Contains 194 million eventualities and 64 million edges.

Demonstrates high quality through intrinsic and extrinsic evaluations.

Supports improved language understanding tasks.

Abstract

Understanding human's language requires complex world knowledge. However, existing large-scale knowledge graphs mainly focus on knowledge about entities while ignoring knowledge about activities, states, or events, which are used to describe how entities or things act in the real world. To fill this gap, we develop ASER (activities, states, events, and their relations), a large-scale eventuality knowledge graph extracted from more than 11-billion-token unstructured textual data. ASER contains 15 relation types belonging to five categories, 194-million unique eventualities, and 64-million unique edges among them. Both intrinsic and extrinsic evaluations demonstrate the quality and effectiveness of ASER.