HelpThing:Work/Recommendations
From WikiThing
Links to this page are automatically created by the main LibraryThing pages.
The help found on these pages is compiled by users just like you.
View a list of all of the HelpThing topics.
While the Library Suggester provides recommendations based on your entire library, this page provides recommendations on a book-by-book basis. It's similar to Amazon's "People who bought X also bought Y", but 1) looks at ownership (or readership) patterns, not just purchasing, and 2) takes into account the wealth of data provided by member-applied tags and Library of Congress subject headings.
Recommendations appear as Number. Title by Author - clicking on the links will take you to the work or author page, respectively.
Contents |
[edit] Search for other books
Entering another title here will take you a list of search results in the Book Suggester, and from there to this page for another book.
[edit] LibraryThing combined recommendations
Compiles the recommendations from each of the various methods of calculation below - books that appear on more than one sub-list will have a higher ranking on the combined list.
[edit] Books with similar tags
This algorithm compares books not on the basis of similar ownership, but on the basis of similar tags, so you can get recommendations based more specifically on subject matter. It uses the most common tags applied to a work by all LibraryThing members, so even if you don't tag your books, it will still generate recommendations. It also weights based on tag obscurity - so the fact that two books share the tag "fiction" counts less towards making them similar than if they share the tag "books about books" or "dystopian steampunk".
[edit] People with this book also have... (more common)
This algorithm generates recommendations based on patterns of book co-occurrence in libraries.
For example, let's say there are 1000 LT members. 500 of them have book X, and 100 of them have book Y. If these two books were unrelated and entirely randomly distributed, you would expect 50 members to have both. That is, if everything is random, 1/2 of any subset of LT members should have book X, so if the subset is "members with book Y", then we would expect the overlap to be 1/2 * 100 = 50. If significantly more than 50 people own both X and Y, that indicates that ownership is not random, and that people interested in one book would be interested in another.
The numbers after each recommendation are the number of people who own both works over the total number of people who own the second work.
[edit] People with this book also have... (more obscure)
This uses a similar algorithm to the one above, but limits it to less-common books.
The numbers after each recommendation are the number of member libraries where the two books are expected to co-occur if distribution is random (50 in the example above), followed by the number of times they actually co-occur.
[edit] Books with similar library subjects and classifications...
This algorithm uses what LibraryThing knows about library-assigned subjects (mostly Library of Congress Subject Headings), Library of Congress Classifications (LCC) and the Dewey Decimal Classifications (DDC) to generate recommendations. Like the "similar tags" recommendations, this algorithm will give you books in the same genre and on the same topic, as well as books that are likely to be shelved together. After each recommendation it lists what types of information were used to make the suggestion.