Collaborative filtering (CF) algorithms predict interests of an active user in order to deal with the overload of information.
Usually, changes of her interests have been ignored in traditional algorithms, which take user’s interest as static data and
product rating in different phase with same weight. So when users’ interests have changed as time goes on, unneeded items
may be recommended. In order to solve above problem, we propose a new item-based collaborative filtering algorithm in this
paper. In this algorithm, named PFCF, we firstly divide users’ rating history into several periods, then users’ interests
distributing in these periods are analyzed by a phrased forecast method, which is used to find user’s different type interests.
The proposed algorithm is strictly tested on the MovieLens data set. The experimental results show its good precision against
other traditional item-based collaborative filtering algorithms.
[Show abstract][Hide abstract] ABSTRACT: Recommender systems apply knowledge discovery techniques to the problem of making personalized recom- mendations for information, products or services during a live interaction. These systems, especially the k-nearest neighbor collaborative filtering based ones, are achieving widespread success on the Web. The tremendous growth in the amount of available information and the number of visitors to Web sites in recent years poses some key challenges for recommender systems. These are: producing high quality recommendations, performing many recommendations per second for millions of users and items and achieving high coverage in the face of data sparsity. In traditional collaborative filtering systems the amount of work increases with the number of participants in the system. New recommender system technologies are needed that can quickly produce high quality recommendations, even for very large-scale problems. To address these issues we have explored item-based collaborative filtering techniques. Item- based techniques first analyze the user-item matrix to identify relationships between different items, and then use these relationships to indirectly compute recommendations for users. In this paper we analyze different item-based recommendation generation algorithms. We look into different techniques for computing item-item similarities (e.g., item-item correlation vs. cosine similarities between item vec- tors) and different techniques for obtaining recommendations from them (e.g., weighted sum vs. regression model). Finally, we experimentally evaluate our results and compare them to the basic k-nearest neighbor approach. Our experiments suggest that item-based algorithms provide dramatically better performance than user-based algorithms, while at the same time providing better quality than the best available user-based algorithms.
[Show abstract][Hide abstract] ABSTRACT: Recommendation algorithms are best known for their use on e-commerce Web sites, where they use input about a customer's interests to generate a list of recommended items. Many applications use only the items that customers purchase and explicitly rate to represent their interests, but they can also use other attributes, including items viewed, demographic data, subject interests, and favorite artists. At Amazon.com, we use recommendation algorithms to personalize the online store for each customer. The store radically changes based on customer interests, showing programming titles to a software engineer and baby toys to a new mother. There are three common approaches to solving the recommendation problem: traditional collaborative filtering, cluster models, and search-based methods. Here, we compare these methods with our algorithm, which we call item-to-item collaborative filtering. Unlike traditional collaborative filtering, our algorithm's online computation scales independently of the number of customers and number of items in the product catalog. Our algorithm produces recommendations in real-time, scales to massive data sets, and generates high quality recommendations.
[Show abstract][Hide abstract] ABSTRACT: Collaborative flltering is regarded as one of the most promising recommendation algorithms. Traditional approaches for collaborative flltering do not take con- cept drift into account. For example, user purchase interests may be volatile. A new mother may be in- terested in baby toys, although previously she had no interest in these. A man may like romantic fllms while he preferred action movies one year ago. Collabora- tive flltering is characterized by concept drift in the real world. To make time-critical predictions, we ar- gue that the target users' recent ratings re∞ect his/her future preferences more than older ratings. In this paper, we present a novel algorithm namely recency- based collaborative flltering to explore the weights for items based on their expected accuracy on the future preferences. Our proposed approach is based on item- based collaborative flltering algorithms. Speciflcally, we design a new similarity function to produce sim- ilarity scores that better re∞ect the reality. Our ex- perimental results have shown that the new algorithm substantially improves the precision of traditional col- laborative flltering algorithms.
Database Technologies 2006, Proceedings of the 17th Australasian Database Conference, ADC 2006, Hobart, Tasmania, Australia, January 16-19 2006; 01/2006
Note: Although carefully collected, accuracy of this list of references cannot be guaranteed.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.