Using CSI Methods to Understand the DNA of Books

Early in our educations we learn the components of a good story. We become familiar with the concept of the characters, settings, plot, and themes. We look for the protagonist, we follow the dialogue and descriptors, we follow the arc of the story and recognize the climax. Some books speak to us as if they were written just for us and when we find a story we love, we share it with our friends or read another book by the same author because we feel a powerful connection with the literature.

Have you ever thought of a book as having DNA? You know, the biological stuff we’re all comprised of and that makes us unique and individual and plays a big part in who and why we are ourselves. Well, Aaron Stanton did just that and then he identified 32,162 genomic measurements and created an ever-expanding database of hundreds of million data points that uniquely classify components of a book.

The result is, a suggestion recommendation engine for books. As Aaron describes it, “If you like vampire books, we will try to identify and suggest other books that are written in a similar fashion and with similar themes that take place at the same period of time or with similar characters, showing which book may be a good fit compared to others you’ve read.  For example, we don’t just look at whether on not the book you liked has “Vampires” in it, but instead that it has 15% vampires, is written like Anne Rice, and also takes place in a modern city.  So we try to find other titles that match that, instead of 50% vampires, written like Stephenie Meyer, taking place around kings and castles.  Both of those might be good books, but we try to find the subtle similarities as well as the obvious.” To take this wealth of knowledge and offer it to readers, Aaron created, which went live only a few short weeks ago and is still growing (and invites you to contribute).

The site is the public-facing component of what began as a back-end programming project Aaron has been working on. “We found that as we described our process to publishers, they always needed to prove our system.  Booklamp is a fun way for us to prove our system and let others have a hand in helping us get better at what we do,” Aaron said.

Both Booklamp and the Book Genome Project that spawned it are financially self-sustaining through back-end tools technology purchased by publishers. This income allows the bibliophiles at Booklamp to offer avid readers an ad-free service that finds other books that they might enjoy.

If you search the database today, you may not find every book you’re looking to DNA-type and connect with other books. That’s because BookLamp is an ambitious project in the early stages and will always be in a state of evolution as new books are published. Currently, the site has DNA fingerprints for only book listings provided by participating publishers, roughly 20,000 titles. But there are big plans ahead and is an ambitious endeavor. About the future, Aaron says, “Attracting publishers to the project to help grow our book database is the primary reason exists. Our biggest criticism of ourselves at this point is that we don’t have enough books to be a “real” project, yet.  That said, I made the resolution some time ago to only read books I found through our tools – because I want to know how it’s working – and I can tell you that 20,000 titles is a fantastic number of books.  It doesn’t feels limiting once you get past the front page and start browsing from book to book.  Where it hurts is when you want to use a book like Ender’s Game as a starting point into the browse and discovery part of the tools, and we don’t have Ender’s Game, yet.  Our goal is to get more books into the system that can serve as a starting place for discovering the books that are here.”

I have to say I was impressed with what this guy has done and I am pretty shocked that publishers are not beating down his door to get involved with such cutting-edge powerful technology. is a truly impressive tool and one that has some unbelievable potential. Good luck, Aaron, can’t wait to see the progress!

Comment (1)

  1. Mark

    Quite ingenious. I can see any bookseller using this tool to improve the upsale. Whattayabet Amazon buys them up?


Leave a Comment

Your email address will not be published. Required fields are marked *