Academic
Publications
Item-based top- N recommendation algorithms

Item-based top- N recommendation algorithms,10.1145/963770.963776,ACM Transactions on Information Systems,Mukund Deshpande,George Karypis

Item-based top- N recommendation algorithms   (Citations: 315)
BibTex | RIS | RefWorks Download
The explosive growth of the world-wide-web and the emergence of e-commerce has led to the development of recommender systems---a personalized information filtering technology used to identify a set of items that will be of interest to a certain user. User-based collaborative filtering is the most successful technology for building recommender systems to date and is extensively used in many commercial recommender systems. Unfortunately, the computational complexity of these methods grows linearly with the number of customers, which in typical commercial applications can be several millions. To address these scalability concerns model-based recommendation techniques have been developed. These techniques analyze the user--item matrix to discover relations between the different items and use these relations to compute the list of recommendations.In this article, we present one such class of model-based recommendation algorithms that first determines the similarities between the various items and then uses them to identify the set of items to be recommended. The key steps in this class of algorithms are (i) the method used to compute the similarity between the items, and (ii) the method used to combine these similarities in order to compute the similarity between a basket of items and a candidate recommender item. Our experimental evaluation on eight real datasets shows that these item-based algorithms are up to two orders of magnitude faster than the traditional user-neighborhood based recommender systems and provide recommendations with comparable or better quality.
Journal: ACM Transactions on Information Systems - TOIS , vol. 22, no. 1, pp. 143-177, 2004
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...One of the main approaches to Collaborative Filtering, adopted in [8,15], relies on the computation of similarity indices among items and on using them for prediction of user likely preferences...
    • ...approach we consider is that of item-based recommendations, which has proved effective in practice(seeforexample[8,15,24]or[1]foramoregeneralsurvey).Ourmaingoalhereisto carry this approach, appropriately adapted, over to the fully decentralized, stigmergy-based scenario we envision...
    • ...5.4). As to the second issue, we compared the performance of our solution with a standard centralized method [15]...
    • ...The algorithm was implemented adopting the optimizations suggested in [15] and tuning parameters for best performance under the data sets we consider...

    Luca Becchettiet al. Recommending items in pervasive scenarios: models and experimental ana...

    • ...In such cases, the problem of finding the best item is usually transformed into the task of recommending to an active user ua a list L(ua) containing N items likely to interest him or her [18, 45]...
    • ...A drawbackof this task is that all itemsof a recommendation listL(u)are considered equally interesting to user u. An alternative setting, described in [18], consists in learning a function L that maps each user u to a list L(u) where items are ordered by their “interestingness” to u. If the test set is built by randomly selecting, for each user u, a single item iu of Iu, the performance of L can be evaluated with the Average Reciprocal ...
    • ...Unlike content-based approaches, which use the content of items previously rated by a user u, collaborative (or social) filtering approaches [18, 31, 41, 47, 60, 45, 70] rely on the ratings of u as well as those of other users in the system...
    • ...Following [1, 5, 10, 18], collaborative filtering methods can be grouped in the two general classes of neighborhood and model-based methods...
    • ...In neighborhoodbased (memory-based [10] or heuristic-based [1]) collaborative filtering [17, 18, 31, 41, 47, 54, 60, 45, 70], the user-item ratings stored in the system are directly used to predict ratings for new items...
    • ...Iuv, are most correlated to those of u. Item-based approaches [18, 47, 45], on the other hand, predict the rating of a user u for an item i based on the ratings of u for items similar to i. In such approaches, two items are similar if several users of the system have rated these items in a similar fashion...
    • ...While user-based methods rely on the opinion of like-minded users to predict a rating, item-based approaches [18, 47, 45] look at ratings given to similar items...
    • ...A common solution for these problems is to fill the missing ratings with default values [10, 18], such as the middle value of the rating range, and the average user or item rating...

    Christian Desrosierset al. A Comprehensive Survey of Neighborhood-based Recommendation Methods

    • ...It is known to provide high quality top-k results [5,12]...
    • ...Computing recommendations from scratch for active users can be prohibitively expensive so CF-based recommender systems typically precompute item-wise similarities and store them in a manner that facilitates quick retrieval [5]...
    • ...The number of users n is typically much larger than the number of items m. Item-based CF was rst proposed in [5, 12] to overcome these limitations...

    Mohammad Khabbazet al. TopRecs: Top-k algorithms for item-based collaborative filtering

    • ...This representation usually leverages item-item similarities and leads to item-based CF algorithms [3]...
    • ...For item-based CF applied to the resource prediction problem, the algorithm suggested by [3] computes the interestingness score of a given user u for a particular resource r as the averaged sum of similarities between r and its neighboring resources Nr that co-occur with u, i.e.,...

    Leandro Balby Marinhoet al. Social Tagging Recommender Systems

    • ...Given an item, these lists help find related items (see [6] for a survey of algorithms)...

    Joseph A. Calandrinoet al. "You Might Also Like:" Privacy Risks of Collaborative Filtering

Sort by: