Event-Arguments Extraction Corpus and Modeling using BERT for Arabic

TL;DR

This paper introduces a new Arabic event-argument corpus, proposes a BERT-based relation extraction method as text entailment, and develops an end-to-end event-arguments extraction system, demonstrating high accuracy and generalization.

Contribution

It provides the first large-scale Arabic event-argument corpus and a novel BERT-based approach for event relation extraction as text entailment.

Findings

Achieved 94.01% F1-score on in-domain data

Achieved 83.59% F1-score on out-of-domain data

Developed an end-to-end event-arguments extraction system

Abstract

Event-argument extraction is a challenging task, particularly in Arabic due to sparse linguistic resources. To fill this gap, we introduce the \hadath corpus ($550$k tokens) as an extension of Wojood, enriched with event-argument annotations. We used three types of event arguments: $agent$, $location$, and $date$, which we annotated as relation types. Our inter-annotator agreement evaluation resulted in $82.23\%$ $Kappa$ score and $87.2\%$ $F_1$-score. Additionally, we propose a novel method for event relation extraction using BERT, in which we treat the task as text entailment. This method achieves an $F_1$-score of $94.01\%$. To further evaluate the generalization of our proposed method, we collected and annotated another out-of-domain corpus (about $80$k tokens) called \testNLI and used it as a second test set, on which our approach achieved promising results ($83.59\%$ $F_1$-score). Last but not least, we propose an end-to-end system for event-arguments extraction. This system is implemented as part of SinaTools, and both corpora are publicly available at {\small \url{https://sina.birzeit.edu/wojood}}