Modality and Negation in Event Extraction

TL;DR

This paper introduces a lexicon-based event extraction system that effectively captures modality in language, improving accuracy in identifying actual versus possible events for applications like QA and fact-checking.

Contribution

The paper presents a novel open-domain system that incorporates modality understanding into event extraction, addressing a key challenge in NLP.

Findings

System effectively captures various modality types

Improves event extraction accuracy in political news domain

Suitable for downstream applications like QA and fact-checking

Abstract

Language provides speakers with a rich system of modality for expressing thoughts about events, without being committed to their actual occurrence. Modality is commonly used in the political news domain, where both actual and possible courses of events are discussed. NLP systems struggle with these semantic phenomena, often incorrectly extracting events which did not happen, which can lead to issues in downstream applications. We present an open-domain, lexicon-based event extraction system that captures various types of modality. This information is valuable for Question Answering, Knowledge Graph construction and Fact-checking tasks, and our evaluation shows that the system is sufficiently strong to be used in downstream applications.