 |
ForumHome | WebHome | Blog | Wiki | MyAccount | Memberlist | Search | Calendar | FAQ | LOGOUT |  |
 |
| Register | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
|
|||||||
Personal tools
|
|
ViewsReadingFrom IPL Wiki
[edit] About Reading RecommendationsMaking book recommendations to patrons has always been an important part of a librarian's job and will likely continue to be so in the future. However, librarians now have a number of tools and systems to supplement their efforts, mainly through modern technologies called recommender systems and techniques like collaborative filtering. Recommender systems and collaborative filtering technologies have actually been around since the early 1990s when GroupLens - one of the first known recommender systems - was created by researchers at the University of Minnesota to recommend UseNet articles to interested readers. Today, the GroupLens technology is being used to recommend bike paths, technology news articles, and movies. Other corporations and research groups have also developed their own collaborative filtering technologies. For anyone who has ever looked at a product on Amazon.com and noticed the section titled "Customers Who Bought This Item Also Bought...", you've seen a collaborative filtering recommender system at work. In recent years, several different types of recommender systems have been launched, providing recommendations for movies (Netflix), music (Pandora) and even potential friends (Facebook) or dating partners (eHarmony). Not surprisingly, these technologies are also being used to help readers keep track of the books they read, and to make personalized book recommendations: And, although they don't actually offer recommendations per se, there are a number of other sites that are helping to change the ways people learn about and find books they want to read:
[edit] How do Reading Recommendations work?Recommender systems can work in a variety of ways, but the most popular (and successful) technique is called "collaborative filtering" [1]. Without getting too technical, here's how collaborative filtering generally works:
While this process seems relatively simple, it actually involves fine-tuning and tweaking several complex calculations and algorithms. Even the most successful algorithms required a great deal of research to create, and even then there's still no guarantee that the recommendations will be accurate. For instance, just last year, Netflix famously awarded $1 million to a team of researchers from AT&T Labs as part of a contest called The Netflix Prize that challenged researchers from around the world to improve Netflix's Cinematch algorithm by 10%. A second contest - Netflix Prize 2 - is set to launch later this year. [edit] Best PracticesSeveral libraries are using LibraryThing:
In addition, "LibraryThing for Libraries" is a simple module libraries can add to their existing OPAC to bring the power of Library 2.0 to their own collection. Here is a list of libraries who utilize this service: http://www.librarything.com/wiki/index.php/LTFL:Libraries_using_LibraryThing_for_Libraries a list of libraries Libraries using GoodReads:
[edit] IPL examples
[edit] Hands on ActivitySign up for a free account on LibraryThing: http://www.librarything.com/ Once you've created your account, find out a little bit more about LibraryThing's features by taking their official tour. Next, edit your profile and add as much (or as little) information as you want. You may want to just start with your first name and your hometown at first. Activity #1 Think of your favorite book (or one you've recently read) and add it to your LibraryThing bookshelf. First go to Add Book and search using the title of the book, the author, the ISBN number, or just some general keywords. To add the book to your collection, simply click on its title. You can also add some tags to the book so that you can find it easier later on. Feel free to add a few more books to your collection in this manner. Activity #2 Visit the IPL's LibraryThing profile. At the bottom of this page, you can see a list of books that the IPL has recently added to its collection. Find one book that sounds interesting and click on its title to find more information about the book. At the top of this page, you'll see some statistics about the book's popularity on LibraryThing along with some of the most popular keywords (or tags) used to describe it. At the bottom, you can see a list of similar books. Click the "Add to your Library" button to add this book to your collection. Keep adding more books to your collection using either of the two strategies described in Activities 1 and 2. Once you add a few books to your collection, you'll notice on your own profile page that LibraryThing offers a list of people with similar collections. You can use these suggestions to find other books you might like. Activity #3 Keep up to date with the IPL's LibraryThing page by adding us as a friend and subscribing to our LibraryThing RSS feed (for more on RSS, check out Thing #10: RSS). Also, feel free to post a comment on our LibraryThing profile to tell us how much you enjoy using LibraryThing, or about some books you think we should add to our profile. Finally, watch this video from YouTube user "annalaurab" on how to use LibraryThing in your library: http://www.youtube.com/watch?v=c0uu_A9JUAU [edit] Articles about Reading Recommendations (and other Recommender Systems)Chen, C., & Chen, A. (2007). Using data mining technology to provide a recommendation service in the digital library. The Electronic Library, 25(6), 711-724. Cooke, N. A. (2008). Social networking in libraries: New tricks of the trade, part II. Public Services Quarterly, 4(4), 353-365. Goldberg, D., D. Nichols, B. M. Oki, and D. Terry. (1992). Using collaborative filtering to weave an information tapestry. Communications of the ACM, 35(12), 61-70. Herlocker, J. L., Konstan, J. A., Terveen, L. G., and Riedl, J. T. (2004). Evaluating collaborative filtering recommender systems. ACM Transactions on Information Systems, 22(1), 5-53. Konstan, J. A., Miller, B. N., Maltz, D., Herlocker, J. L., Gordon, L. R., and Riedl, J. 1997. GroupLens: applying collaborative filtering to Usenet news. Communications of the ACM, 40(3), 77-87. Monnich, M., & Spiering, M. (2008). Adding value to the library catalog by implementing a recommendation system. D-Lib Magazine, 14(5-6). Retrieved from http://www.dlib.org/dlib/may08/monnich/05monnich.html. Resnick P, Iacovou N., Suchak M., Bergstrom, and Riedl J. (1994). GroupLens: An open architecture for collaborative filtering of netnews. In Proceedings of ACM 1994 Conference on Computer Supported Cooperative Work (CSCW '94), 22-26 October 1994 (pp. 175-186). Chapel Hill, NC: ACM. Sarwar, B., Karypis, G., Konstan, J., and Reidl, J. (2001). Item-based collaborative filtering recommendation algorithms. In Proceedings of the 10th International Conference on World Wide Web (WWW '01), 1-5 May 2001 (pp. 285-295). Hong Kong, Hong Kong: ACM. Whitney, C., & Schiff, L. (2006). The Melvyl Recommender Project: Developing Library Recommendation Services. D-Lib Magazine, 12(12). Retrieved from http://www.dlib.org/dlib/december06/whitney/12whitney.html. [edit] Feedback and Prize DrawingTo give feedback to the IPL about the 15 Things and to register for the prize drawing, please visit: http://vll.ipl.org/15things/index.html
Return to Main Page |
