Local DiskSocial Information Filtering: Algorithms for Automating "Word of Mouth''Social Information Filtering: Algorithms for Automating "Word of Mouth''CHI '95 ProceedingsTopIndexes PapersTOC Social Information Filtering: Algorithms for Automating "Word of Mouth''Upendra Shardanand and Pattie Maes MIT Media-Lab20 Ames Street Rm. 305Cambridge, MA [email protected]\\ (617) 253-7441 [email protected](617) 253-7442 © ACM AbstractThis paper describes a technique for making personalized recommendations from any type of database to a user based on similarities between the interest profile of that user and those of other users. In particular, we discuss the implementation of a networked system called Ringo, which makes personalized recommendations for music albums and artists. Ringo's database of users and artists grows dynamically as more people use the system and enter more information. Four different algorithms for making recommendations by using social information filtering were tested and compared. We present quantitative and qualitative results obtained from the use of Ringo by more than 2000 people. Keywords:social information filtering, personalized recommendation systems, user modeling, information retrieval, intelligent systems, CSCW. IntroductionRecent years have seen the explosive growth of the sheer volume of information. The number of books, file:///C|/Documents%20and%20Settings/Administrator.CAMRY.000/Desktop/shardanand_1995.html (1 of 15)9/18/2005 8:59:48 AMSocial Information Filtering: Algorithms for Automating "Word of Mouth''movies, news, advertisements, and in particular on-line information, is staggering. The volume of things is considerably more than any person can possibly filter through in order to find the ones that he or she will like. People handle this information overload through their own effort, the effort of others and some blind luck. First of all, most items and information are removed from the stream simply because they are either inaccessible or invisible to the user. Second, a large amount of filtering is done for us. Newspaper editors select what articles their readers want to read. Bookstores decide what books to carry. However with the dawn of the electronic information age, this barrier will become less and less a factor. Finally, we rely on friends and other people whose judgement we trust to make recommendations to us. We need technology to help us wade through all the information to find the items we really want and need, and to rid us of the things we do not want to be bothered with. The common and obvious approach used to tackle the problem of information filtering is content-based filtering[1]. Keyword-based filtering and patent semantic indexing [2] are some example content-based filtering techniques. Content-based filtering techniques recommend items for the user's consumption based on correlations between the content of the items and the user's preferences. For example, the system may try to correlate the presence of keywords in an article with the user's taste. However, content-based filtering has limitations: ● Either the items must be of some machine parsable form (e.g. text), or attributes must have been assigned to the items by hand. With current technology, media such as sound, photographs, art, video or physical items cannot be analyzed automatically for relevant attribute information. Often it is not practical or possible to assign attributes by hand due to limitations of resources. ● Content-based filtering techniques have no inherent method for generating serendipitous finds. The system recommends more of what the user already has seen before (and indicated liking). In practice, additional hacks are often added to introduce some element of serendipity. ● Content-based filtering methods cannot filter items based on quality, style or point-of-view. For example, they cannot distinguish between a well written an a badly written article if the two articles happen to use the same terms. A complementary filtering technique is needed to address these issues. This paper presents social information filtering, a general approach to personalized information filtering. Social Information filtering essentially automates the process of ``word-of-mouth'' recommendations: items are recommended to a user based upon values assigned by other people with similar taste. The system determines which users have similar taste via standard formulas for computing statistical correlations. Social Information filtering overcomes some of the limitations of content-based filtering. Items being filtered need not be amenable to parsing by a computer. Furthermore, the system may recommend items to the user which are very different (content-wise) from what the user has indicated liking before. Finally, recommendations are based on the quality of items, rather than more objective properties of the items themselves. file:///C|/Documents%20and%20Settings/Administrator.CAMRY.000/Desktop/shardanand_1995.html (2 of 15)9/18/2005 8:59:48 AMSocial Information Filtering: Algorithms for Automating "Word of Mouth''This paper details the implementation of a social information filtering system called Ringo, which makes personalized music recommendations to people on the Internet. Results based on the use of this system by thousands of actual users are presented. Various social information filtering algorithms are described, analyzed and compared. These results demonstrate the strength of social information filtering and its potential for immediate application. RINGO: A PERSONALIZED MUSICRECOMMENDATION SYSTEMSocial Information filtering exploits similarities between the tastes of different users to recommend (or advise against) items. It relies on the fact that people's tastes are not randomly distributed: there are general trends and patterns within the taste of a person and as well as between groups of people. Social Information filtering automates a process of ``word-of-mouth'' recommendations. A significant difference is that instead of having to ask a couple friends about a few items, a social information filtering system can consider thousands of other people, and consider thousands of different items, all happening autonomously and automatically. The basic idea is: 1. The system maintains a user profile, a record of the user's interests (positive as well as negative) in specific items. 2. It compares this profile to the profiles of other users, and