Data-driven Coreference-based Ontology Building

TL;DR

This paper presents a data-driven method for constructing biomedical ontologies by analyzing coreference chains in a large corpus, using graph analysis to identify hierarchical and conceptual relationships.

Contribution

It introduces a novel approach that leverages coreference resolution and graph centrality to automatically build and refine biomedical ontologies from text data.

Findings

Generated a biomedical ontology with significant overlap to existing human-curated ontologies.

Demonstrated the effectiveness of betweenness centrality in identifying hierarchical relations.

Provided publicly available coreference chains and ontology data for further research.

Abstract

While coreference resolution is traditionally used as a component in individual document understanding, in this work we take a more global view and explore what can we learn about a domain from the set of all document-level coreference relations that are present in a large corpus. We derive coreference chains from a corpus of 30 million biomedical abstracts and construct a graph based on the string phrases within these chains, establishing connections between phrases if they co-occur within the same coreference chain. We then use the graph structure and the betweeness centrality measure to distinguish between edges denoting hierarchy, identity and noise, assign directionality to edges denoting hierarchy, and split nodes (strings) that correspond to multiple distinct concepts. The result is a rich, data-driven ontology over concepts in the biomedical domain, parts of which overlaps significantly with human-authored ontologies. We release the coreference chains and resulting ontology under a creative-commons license, along with the code.