There are a number of major awards are given to the ‘best’ new Science and Fiction fantasy novels each year, such as the Hugo, Nebula, and World Fantasy awards. Predicting which books might win is made especially difficult by the paucity of publicly available sales and review data as a function of time.

In this project we constructed a predictive model to select award winners from a pool of nominees based on publicly available information about the books and a proxy topicality score describing world state and zeitgeist for the year of publication.