N24News: A New Dataset for Multimodal News Classification
This addresses the lack of multimodal datasets for news classification, offering a resource for researchers, but it is incremental as it builds on existing multimodal methods in a specific domain.
The authors tackled the problem of news classification by creating N24News, a multimodal dataset with text and images from the New York Times across 24 categories, and found that multimodal classification outperforms text-only methods, increasing accuracy by up to 8.11% depending on text length.
Current news datasets merely focus on text features on the news and rarely leverage the feature of images, excluding numerous essential features for news classification. In this paper, we propose a new dataset, N24News, which is generated from New York Times with 24 categories and contains both text and image information in each news. We use a multitask multimodal method and the experimental results show multimodal news classification performs better than text-only news classification. Depending on the length of the text, the classification accuracy can be increased by up to 8.11%. Our research reveals the relationship between the performance of a multimodal classifier and its sub-classifiers, and also the possible improvements when applying multimodal in news classification. N24News is shown to have great potential to prompt the multimodal news studies.