Information, Medium & Society: Eighteenth International Conference on Publishing Studies

Abstract

What’s in your collection? A common challenge faced by publishers, librarians, and all those responsible for content curation in the digital age can be summarized in two words: finding content. Metadata and resource discovery can be frustrating, as classifying information across multiple databases, platforms and portals is by nature inconsistent and incomplete. Most commonly, content is sorted and classified by a set of keywords, but this comes up short because keywords become more or less important depending on the context in which they are used. For example: a very small portion of a book or scholarly article may be quite important in relation to a second theme or concept. However, because it is not the main theme, keyword categorization would miss this text. What if there were a better solution? Artificial Intelligence offers the ability to ingest and read digital collections in their entirety, categorizing them on-demand according to the curator or end-user’s needs. In this paper, Manisha Bolina of Yewno, a leading provider of next generation discovery tools, explains how AI can power concept-based discovery for publishers, librarians, and researchers. She is joined by Paolo Sirito, of Università Cattolica del Sacro Cuore who considers Cattolica’s use case in which Yewno helped his university librarians and researchers to visualise, discover, and better understand their own content corpus.