ClimaText: A Dataset for Climate Change Topic Detection

TL;DR

ClimaText introduces a new dataset for climate change topic detection in text, highlighting the limitations of keyword models and the potential of context-based algorithms like BERT for complex, implicit climate-related content.

Contribution

This paper presents ClimaText, a publicly available dataset for sentence-level climate change detection and evaluates different models, emphasizing the need for improved methods.

Findings

Keyword models are inadequate for complex climate topics

BERT can identify implicit and complex climate-related patterns

There is significant potential for improving climate change detection methods

Abstract

Climate change communication in the mass media and other textual sources may affect and shape public perception. Extracting climate change information from these sources is an important task, e.g., for filtering content and e-discovery, sentiment analysis, automatic summarization, question-answering, and fact-checking. However, automating this process is a challenge, as climate change is a complex, fast-moving, and often ambiguous topic with scarce resources for popular text-based AI tasks. In this paper, we introduce \textsc{ClimaText}, a dataset for sentence-based climate change topic detection, which we make publicly available. We explore different approaches to identify the climate change topic in various text sources. We find that popular keyword-based models are not adequate for such a complex and evolving task. Context-based algorithms like BERT \cite{devlin2018bert} can detect, in addition to many trivial cases, a variety of complex and implicit topic patterns. Nevertheless, our analysis reveals a great potential for improvement in several directions, such as, e.g., capturing the discussion on indirect effects of climate change. Hence, we hope this work can serve as a good starting point for further research on this topic.