Beyond the Book
The ultimate aim of this project is to design and develop methods and tools for classifying attributes of books such as genre, style, tone, and likelihood of being popular. Towards this end we will make use of various information types available on books and users of the Kobo catalog, including the text, meta-data associated with the text, and user features associated with readers of the text. This is a large undertaking. As a first step, the intern and research team will tackle the problem of genre classification, while keeping in mind that the task is one among a collection of desired automatic tagging tools for books. Through the use of NLP and machine learning techniques, books can be categorized and referred to readers via interests that they have expressed. This project is likely to provide valuable insight into both books and readers, allowing Kobo to provide users with higher quality recommendations, interesting reading lists, and deeper understanding of both books and users.