Just use the box on the home page - all you need is your email address! Or, if you start typing the name or author of a book you like on the home page then click on the orange link which matches it best, at the bottom of the results page there's a box where you can register and your original book will be the first one on your new list.
We add new features to the site every few months and have lots more plans in the pipeline! You can keep up-to-date with what's going on via our Twitter feed and Facebook page, or by signing up at the site and joining our mailing list.
WSIRN produces recommendations based purely on collective taste: when books are entered into the same favorites list, they become associated with each other. The more often particular items appear on different lists, the stronger that association becomes. Purely and simply, WSIRN represents mass opinion about books. For the technically minded, this is a collaborative filtering system, using our own bespoke algorithm called 'Incidence Bias Weighting' and partly using association rules.
If you simply want a quick recommendation, you only need to enter one book; click on the best match in the list that appears as you type and you will get some recommendations. The best thing to do, though, is enter a list of favorites. This will give you more focused recommendations, as well as helping to build the database and share your taste with other users of the site. If you enter your email address, your list of favorites will be remembered next time you visit, so you can then add new titles, create more lists, and get even more appropriate recommendations.
Aside from the warm, fuzzy feeling of contributing to the community, there are lots of ways you can use lists, now that we've added the multiple list feature: you can keep track of books you've read and liked, or have a wish list of things you want to read, or even a list of books you hated. You could also build lists of what you think are the 'top 10 psychological thrillers' or whatever and use the new share feature (under each list) to let other people see it. You will also get more personalised recommendations if you select multiple books in your list before hitting the 'What Should I read next?' button.
We filter these out because it seems obvious that you'd be interested in the other books by that author. You can always enter their name in the search box to see which of their books are in our database. The database is updated every day, so come back soon if you don't see them all!
At the moment anyone can add books by ISBN number (if you register for free, you can add multiple ISBNs at once as a comma-separated list) - our system can usually then fetch the info. But if you want your book/s to be found in recommendations, they need to be in users' book lists - you can kick-start the process by building lists of similar titles to your own (including your own, of course). We are working on developing tools to help authors and publishers more directly.
No recommendation algorithm is perfect, but we're constantly tidying the database to improve things - sometimes there are entries which are spelled wrongly from the original source database. The occasional wacky suggestion is usually down to the search book not appearing in many users' lists. But the results improve all the time, so do come back and try again!
Most of the time the database will identify your book by title/author straight away (you will see a list of close matches, with the most likely one at the top). Now and again it may not come up with what you're looking for immediately. If so, try checking you've spelt them correctly - also, try just entering the author's surname or a keyword from the book title. If that still doesn't work, you can enter the ISBN number from the book's bar code. We can then look that up elsewhere. If you're really stuck, get in touch with us via the links below and we'll try to help you out as soon as we can.
We're doing our best to iron these out - it's a problem that all the big book sites face. Some other sites like to have as many different editions as possible - but we think focusing on the 'work' is more important for providing meaningful recommendations. We are gradually removing cases where multiple editions confuse matters.
Amazon certainly offers a useful recommendation system, but it is based on your past buying patterns and those of other users - however, you don't always buy items for yourself, do you? Our system has no commercial agenda at all (though you can buy the recommended items through Amazon, of course), and is purely based on items that real people actually like. We're not trying to urge you to buy particular bestsellers or anything like that - we simply want to help people share their favorite items with each other.
As an Amazon Associate we earn from qualifying purchases (Amazon and the Amazon logo are trademarks of Amazon.com, Inc. or its affiliates). This helps to cover our site hosting and programming costs.