Back in Spring 2016 we announced that Altmetric was now tracking attention to books and their individual chapters based on the ISBN and other unique identifiers. This was a really big step towards helping authors of monographs track and evidence the broader engagement relating to their research.
Tracking attention for books is not an easy task, and we’re still learning more all the time about how we can improve! The metadata is often messy and the way that people talk about them online is very different to the way they discuss and share journal articles.
Despite that, we’ve been making improvements along the way. We did some initial work last year to bring in book reviews from selected sources, and have integrated Open Syllabus Project data to showcase valuable insights, such as the ones described by Michigan Publishing in this video.
ONIX to the rescue
As we’ve set up book tracking for more publishers, the need for a more reliable source of book metadata has become even more pressing. Even though this metadata is pretty simple – title, author, year of publication, ISBN – we use it in all sorts of contexts to improve matching for authors and publishers. Luckily, a pretty good solution exists, and with a bit of work we’re now able to ingest the ONIX feed of any publisher, and we started with the MIT Press.
What is an ONIX feed?
An ONIX feed is an XML metadata structure style and not a specific software. A feed that uses the ONIX structure conforms to the EDitEUR standards of data display. It is intended to be a rich collection of data related to a publication, most often books and e-books. This data includes things like; publisher name, publication date, and cover image.
Unlike with other types Altmetric tracking, the ONIX feed sends us the metadata for an output. Instead of going to the publication page and scraping the data, the ONIX feed “pushes” the information to us.
Many publishers already use an ONIX feed to send the bibliographic information about their books to platforms such as Amazon.
What’s the real benefit?
Receiving information directly from the publisher doesn’t necessarily mean that we will find more attention for a book, but it does mean that we can provide a better record of the book to users. Here’s a details page where you can see this in action:
By ingesting the ONIX feed we and the publisher can be sure that the book information is correct on the Altmetric details page, and also easily display the book cover – making it a much brighter user experience!
We use a variety of text mining approaches to find mentions of journal articles in news stories and policy documents. These are made possible by databases like Pubmed and CrossRef, where we know we can find reliable metadata. By building up a similar database for books by working with publishers directly we hope that over the coming months we can improve mentions of books in those sources too.
Want this for your books?
We’d be happy to chat! Get in touch at email@example.com to discuss how we can provide these valuable insights for your books.