Interactive Topic Model for the Signs Journal: Data

In this study, the corpus that was modeled included all of the articles from the Signs journal (1975-2014), resulting in a 70-topic model. Each article is described as a mixture of verbal patterns or topics. Every topic is a family of words that tend to occur together in articles—as, for example, world, global, and states do. Every article is in turn divided into multiple topics. Words, too, may belong to more than one topic. For example, the word “women” is part of many topics. By following the shifting proportions of topics over time, the study provides ways into thinking about the history of Signs in its first four decades of publishing.