Ontology-Grounded Topic Modeling for Climate Science Research

TL;DR

This paper introduces an ontology-grounded topic modeling approach tailored for climate science research, enhancing topic interpretability and supporting faster research synthesis and discovery.

Contribution

It proposes a novel method that integrates domain ontologies into topic modeling, improving topic quality and interpretability in scientific research domains.

Findings

Enhanced topic coherence and clarity

Faster research understanding and synthesis

Supports automatic ontology generation

Abstract

In scientific disciplines where research findings have a strong impact on society, reducing the amount of time it takes to understand, synthesize and exploit the research is invaluable. Topic modeling is an effective technique for summarizing a collection of documents to find the main themes among them and to classify other documents that have a similar mixture of co-occurring words. We show how grounding a topic model with an ontology, extracted from a glossary of important domain phrases, improves the topics generated and makes them easier to understand. We apply and evaluate this method to the climate science domain. The result improves the topics generated and supports faster research understanding, discovery of social networks among researchers, and automatic ontology generation.