Predicting the Politics of an Image Using Webly Supervised Data
This addresses the challenge of detecting visual political bias in media, which is important for understanding media influence, but is incremental as it builds on existing webly supervised and multimodal learning approaches.
The authors tackled the problem of predicting the political leaning of images from news media using a large-scale dataset of over one million images from left- and right-leaning sources, and developed a two-stage method that outperformed strong baselines.
The news media shape public opinion, and often, the visual bias they contain is evident for human observers. This bias can be inferred from how different media sources portray different subjects or topics. In this paper, we model visual political bias in contemporary media sources at scale, using webly supervised data. We collect a dataset of over one million unique images and associated news articles from left- and right-leaning news sources, and develop a method to predict the image's political leaning. This problem is particularly challenging because of the enormous intra-class visual and semantic diversity of our data. We propose a two-stage method to tackle this problem. In the first stage, the model is forced to learn relevant visual concepts that, when joined with document embeddings computed from articles paired with the images, enable the model to predict bias. In the second stage, we remove the requirement of the text domain and train a visual classifier from the features of the former model. We show this two-stage approach facilitates learning and outperforms several strong baselines. We also present extensive qualitative results demonstrating the nuances of the data.