Yewno: Transforming Information into Knowledge

Yewno: Transforming Information into Knowledge

At Yewno, we describe what we do as “Transforming Information into Knowledge”.  Yewno allows you to surface the information you are looking for, in less time than ever before, from amid the overload of content available today.  Our dynamic knowledge graph connects concepts and surfaces relationships between concepts that might otherwise go unnoticed.  Drawing from our vast and growing repository of scholarly and academic content the possibilities for knowledge discovery are endless. All of our applications are visually engaging, intuitive and fun to use. 

The idea for Yewno came out of research activity in the space of applied Artificial Intelligence.  Having successfully carried out his first proof of concept working with the financial community in London, Yewno CEO Ruggero Gramatica wanted to carry out the second proof of concept where the amount of unstructured data was posing a challenge in understanding the subject matter.  He found the perfect opportunity in Switzerland with a Biotech company that was looking to repurpose existing drug molecules to cure rare diseases.  It was here that he met Professor Michael Keller, Stanford University’s Vice Provost and University Librarian.  Professor Keller invited Ruggero to leverage the work he was doing with artificial intelligence models to build an interdisciplinary inference engine, which became Yewno’s first product:  Yewno Discover.  Today, the Yewno Discover platform transforms the research experience into the creative, inferential discovery that helps researchers, scholars, students, and teachers to surface hidden connections and explore more deeply.

How Yewno Discover Works

Yewno ingests and reads full-text digital content.  We have over two hundred million sources of our own, including books and scholarly articles, and we can also ingest libraries’ own digital repositories.  Once the content is read it is stored in our underlying knowledge graph.  Users access content using Yewno Discover’s visually intuitive and attractive user interface ,then type in the concept that they’d like to research.  The concept is surfaced and the user is given the opportunity to disambiguate the results - i.e. if the concept were “Java” he or she can specify if they are looking for Java the island, Java the computer language, or Java the term for coffee.  A similar example might be Jaguar the car, or Jaguar the animal.  Once this has been done, Yewno Discover displays the concept in a knowledge graph format and shows links to other related ideas.  It is up to the researcher which direction they’d like to go from here, and how deeply they’d like to explore concepts, related concepts or the relationships between these.  Because the system reads and stores full text rather than just keywords, the possibilities for connection grow exponentially, paving the way for novel research. 

Edtech Use Cases

Yewno Discover has gained a dedicated following and has been rapidly expanding within higher education in Universities around the world.  At the University level, Yewno has recently expanded our offering to include Repository Visualization, in which Universities with their own resources such as published research, theses, and dissertations can view their own resources, manage their collections, and connect researchers working on similar projects.  

High schools have also adopted the technology, and the results have been astonishing.  One high school librarian has told us that over the course of just one semester, our students accessed over 275,000 unique documents from 5,080 academic publications and 3,000 dissertations.  All of the content was highly scholarly, pulling from publications such as Proceedings of SPIE--the International Society for Optical Engineering, The Journal of Hellenic Studies, and Publications of the Astronomical Society of the Pacific.   This same librarian says that prior to using Yewno Discover, she often assigned research only to get back the same report time-after-time, drawing from sources such as Wikipedia.