A presentation by Zachary Steinert-Threlkeld (UCLA). This event is part of the Political Economy Lunch Seminar (PELS).
Treating images as data has become increasingly popular in political science. While existing classifiers for images reach high levels of accuracy, it is difficult to systematically assess the visual features on which they base their classification. This paper presents a two-level classification method that addresses this transparency problem. At the first stage, an image segmenter detects the objects present in the image. In the second stage, a non-visual feature vector is created from these objects and used as input for standard machine learning classifiers to discriminate between images. We apply this method to a new dataset of more than 140,000 images to detect which ones display political protest. This analysis demonstrates three advantages to this paper's approach. First, identifying objects in images improves transparency by providing human-understandable labels for the objects shown on an image. Second, knowing these objects, we can analyze what type of objects distinguish protest images from non-protest ones. Third, comparing the importance of objects across countries reveals how protest behavior varies. These insights are not available using conventional computer vision classifiers and provide new opportunities for comparative research. .
Please register for this event by sending an email to Simone Dudziak (dudziak[at]hertie-school[dot]org) or by using the registration form on the sidebar.
