The Bestseller Code: Predicaments of Predicting High Value Fiction
Speaker: Joris Van Zundert (Huygens Institute for the History of the Netherlands)
Publishing literature is a precarious business these days. Ever greater numbers of books are being published, but revenue for literary production is falling consistently over the years. Increasingly literary publishers rely on a few bestsellers to "win big" so publishing literature remains a viable business. Therefore it is not just interesting from a literary studies point of view to know what text immanent properties are characteristic for bestsellers. This knowledge may even prove vital for literary production to survive. At the Huygens Institute in Amsterdam a research group in the literary studies department is taking a computational approach to this problem. Their research applies advanced computational statistics and machine learning techniques to pinpoint textual markers that can be associated with successful fiction writing. In this lecture their methods are explained and results are critically evaluated.
The event will be held online on Zoom, https://unive.zoom.us/j/91596187539
L'evento si terrà in inglese
VeDPH; Dipartimento di Studi Umanistici