English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Objective: Biomedical Knowledge Graphs play a pivotal role in variousbiomedical research domains. Concurrently, term clustering emerges as a crucialstep in constructing these knowledge graphs, aiming to identify synonymousterms. Due to a lack of knowledge, previous contrastive learning models trainedwith Unified Medical Language System (UMLS) synonyms struggle at clusteringdifficult terms and do not generalize well beyond UMLS terms. In this work, weleverage the world knowledge from Large Language Models (LLMs) and proposeContrastive Learning for Representing Terms via Explanations (CoRTEx) toenhance term representation and significantly improves term clustering.Materials and Methods: The model training involves generating explanations fora cleaned subset of UMLS terms using ChatGPT. We employ contrastive learning,considering term and explanation embeddings simultaneously, and progressivelyintroduce hard negative samples. Additionally, a ChatGPT-assisted BIRCHalgorithm is designed for efficient clustering of a new ontology. Results: Weestablished a clustering test set and a hard negative test set, where our modelconsistently achieves the highest F1 score. With CoRTEx embeddings and themodified BIRCH algorithm, we grouped 35,580,932 terms from the BiomedicalInformatics Ontology System (BIOS) into 22,104,559 clusters with O(N) queriesto ChatGPT. Case studies highlight the model's efficacy in handling challengingsamples, aided by information from explanations. Conclusion: By aligning termsto their explanations, CoRTEx demonstrates superior accuracy over benchmarkmodels and robustness beyond its training set, and it is suitable forclustering terms for large-scale biomedical ontologies.

CoRTEx: Contrastive Learning for Representing Terms via Explanations with Applications on Constructing Biomedical Knowledge Graphs