ReOnto: A Neuro-Symbolic Approach for Biomedical Relation Extraction

TL;DR

ReOnto introduces a neuro-symbolic method combining graph neural networks and biomedical ontologies to improve relation extraction accuracy in biomedical texts, outperforming existing baselines.

Contribution

The paper presents ReOnto, a novel neuro-symbolic approach that integrates ontologies with graph neural networks for enhanced biomedical relation extraction.

Findings

ReOnto outperforms baseline methods by approximately 3% on BioRel and ADE datasets.

Using ontologies as prior knowledge improves relation extraction accuracy.

The approach effectively leverages symbolic knowledge to address challenges in biomedical RE.

Abstract

Relation Extraction (RE) is the task of extracting semantic relationships between entities in a sentence and aligning them to relations defined in a vocabulary, which is generally in the form of a Knowledge Graph (KG) or an ontology. Various approaches have been proposed so far to address this task. However, applying these techniques to biomedical text often yields unsatisfactory results because it is hard to infer relations directly from sentences due to the nature of the biomedical relations. To address these issues, we present a novel technique called ReOnto, that makes use of neuro symbolic knowledge for the RE task. ReOnto employs a graph neural network to acquire the sentence representation and leverages publicly accessible ontologies as prior knowledge to identify the sentential relation between two entities. The approach involves extracting the relation path between the two entities from the ontology. We evaluate the effect of using symbolic knowledge from ontologies with graph neural networks. Experimental results on two public biomedical datasets, BioRel and ADE, show that our method outperforms all the baselines (approximately by 3\%).