Book Genome Dataset

Tag Genome is a data structure containing scores indicating the degree to which tags apply to items, such as movies or books. This dataset contains a Tag Genome generated for a set of books along with the data used for its generation (raw data). Raw data includes information, such as titles, authors, user ratings and book-tag ratings. The Tag Genome scores have been generated with two state-of-the-art algorithms: Glmer [Vig et al., 2012] and TagDL [Kotkov et al., 2021]. The code is available in the following GitHub repository: https://github.com/Bionic1251/Revisiting-the-Tag-Relevance-Prediction-Problem