Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Detecting Frames in News Headlines and Lead Images in U.S. Gun Violence Coverage

Isidora Chara Tourni, Lei Guo, Hengchang Hu, Edward Halim, Prakash Ishwar, Taufiq Daryanto, Mona Jalal, Boqi Chen, Margrit Betke, Fabian Zhafransyah, Sha Lai, Derry Tanti Wijaya

TL;DR

This paper tackles the problem of automatically identifying news frames by jointly modeling headlines and lead images in U.S. gun violence coverage. It introduces a multimodal extension to the Gun Violence Frame Corpus that includes lead-image relevance, Subject/Race/Ethnicity annotations, Web Entity tags, image captions, and article summaries, enabling scalable analysis of text-image framing. Through unimodal and multimodal experiments, it shows that image contextual information substantially improves frame prediction when images are relevant, with the best multimodal setup achieving up to $87\%$ accuracy on articles with relevant images and $82.4\%$ on all articles, while frame relevance and frame concreteness are important explanatory factors (Pearson $r\approx0.69$). The work contributes a publicly available dataset and methodological insights for combining multimodal signals in framing, with practical implications for editors and researchers studying media framing of gun violence.

Abstract

News media structure their reporting of events or issues using certain perspectives. When describing an incident involving gun violence, for example, some journalists may focus on mental health or gun regulation, while others may emphasize the discussion of gun rights. Such perspectives are called \say{frames} in communication research. We study, for the first time, the value of combining lead images and their contextual information with text to identify the frame of a given news article. We observe that using multiple modes of information(article- and image-derived features) improves prediction of news frames over any single mode of information when the images are relevant to the frames of the headlines. We also observe that frame image relevance is related to the ease of conveying frames via images, which we call frame concreteness. Additionally, we release the first multimodal news framing dataset related to gun violence in the U.S., curated and annotated by communication researchers. The dataset will allow researchers to further examine the use of multiple information modalities for studying media framing.

Detecting Frames in News Headlines and Lead Images in U.S. Gun Violence Coverage

TL;DR

This paper tackles the problem of automatically identifying news frames by jointly modeling headlines and lead images in U.S. gun violence coverage. It introduces a multimodal extension to the Gun Violence Frame Corpus that includes lead-image relevance, Subject/Race/Ethnicity annotations, Web Entity tags, image captions, and article summaries, enabling scalable analysis of text-image framing. Through unimodal and multimodal experiments, it shows that image contextual information substantially improves frame prediction when images are relevant, with the best multimodal setup achieving up to accuracy on articles with relevant images and on all articles, while frame relevance and frame concreteness are important explanatory factors (Pearson ). The work contributes a publicly available dataset and methodological insights for combining multimodal signals in framing, with practical implications for editors and researchers studying media framing of gun violence.

Abstract

News media structure their reporting of events or issues using certain perspectives. When describing an incident involving gun violence, for example, some journalists may focus on mental health or gun regulation, while others may emphasize the discussion of gun rights. Such perspectives are called \say{frames} in communication research. We study, for the first time, the value of combining lead images and their contextual information with text to identify the frame of a given news article. We observe that using multiple modes of information(article- and image-derived features) improves prediction of news frames over any single mode of information when the images are relevant to the frames of the headlines. We also observe that frame image relevance is related to the ease of conveying frames via images, which we call frame concreteness. Additionally, we release the first multimodal news framing dataset related to gun violence in the U.S., curated and annotated by communication researchers. The dataset will allow researchers to further examine the use of multiple information modalities for studying media framing.

Paper Structure

This paper contains 11 sections, 3 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Dataset
  4. Experiments
  5. Unimodal
  6. Multimodal
  7. Relevance
  8. Discussion of Results
  9. Conclusions
  10. Ethical Considerations
  11. Appendix

Figures (3)

  • Figure 1: Sample images for each frame (from left to right and top to down): 2nd Amendment, Gun Control, Politics, Mental Health, School/Public Space Safety, Race/Ethnicity, Public Opinion, Society/Culture, Economic Consequences.
  • Figure 2: Frame relevance ratio and average concreteness.
  • Figure 3: Per Frame F1 score of the best performing frame prediction methods for (a) All Articles (left), (b) Articles with Relevant Images (right).