Search

With the rapid development of Internet, the amount of information on the Web grows explosively, people often feel puzzled and helpless in finding and getting the information they really need. For overcoming this problem, recommender systems such as singular value decomposition (SVD) method help users finding relevant information, products or services by providing personalized recommendations based on their profiles. SVD is a powerful technique for dimensionality reduction. However, due to its expensive computational requirements and weak performance for large sparse matrices, it has been considered inappropriate for practical applications involving massive data.Thus, to extract information in which the user is interested from a massive amount of data, we propose a personalized recommendation algorithm which is called ApproSVD algorithm based on approximating SVD in this paper. The trick behind our algorithm is to sample some rows of a user-item matrix, rescale each row by an appropriate factor to form a relatively smaller matrix, and then reduce the dimensionality of the smaller matrix. Finally, we present an empirical study to compare the prediction accuracy of our proposed algorithm with that of Drineas's LINEARTIMESVD algorithm and the standard SVD algorithm on MovieLens dataset and Flixster dataset, and show that our method has the best prediction quality. Furthermore, in order to show the superiority of the ApproSVD algorithm, we also conduct an empirical study to compare the prediction accuracy and running time between ApproSVD algorithm and incremental SVD algorithm on MovieLens dataset and Flixster dataset, and demonstrate that our proposed method has better performance overall. [ABSTRACT FROM AUTHOR]

Recommender systems have become one of the necessary tools to help a web user find a potentially interesting resource based on their preferences. In implicit recommender systems, the recommendations are made based on the implicit information of the web users i.e. data collected from web logs or cookies without knowing users preferences. Developing such a recommender system is complex due to the huge amount of anonymous noisy data. In this paper we present a Particle Swarm Optimization (PSO) based clustering approach called Hierarchical Particle Swarm Optimization based clustering (HPSO-clustering) for building a recommender system based on implicit web usage data. The approach mimics multi-agent properties of the particles of a swarm and divide the problem space into smaller sub-spaces i.e. clusters. Each cluster represents a particular group of user with similar interests. Later, the K-nearest neighbours of the most relevant cluster are generated as recommendations for a web user and ranked based on their distance. We performed different experiments for preprocessing, to assess the quality of clusters, and for the accuracy of recommendations. An overall accuracy of 65% to 95% was achieved for different scenarios, while in some cases the accuracy touched 100 precent when the selection was made from the top-5 recommendations. [ABSTRACT FROM AUTHOR]

Recommender systems are common web applications which support users in finding suitable products in large and/or complex product domains. Although state-of-the-art systems manage to accomplish the task of finding and presenting suitable products they show big deficits in their models of human behavior. Time limitations, cognitive capacities and willingness to cognitive effort bound rational decision making which can lead to unforeseen side effects and consequently to sub-optimal decisions. Decoy effects are cognitive phenomena which are omni-present on result pages but state-of-the-art recommender systems are completely unaware of such effects. Due to the fact that such effects constitute one source of irrational decisions their identification and, if necessary, the neutralization of their biasing potential is extremely important. This paper introduces an approach for identifying and minimizing decoy effects on recommender result pages. To support the suggested approach we present the results of a corresponding user study which clearly proves the concept. Moreover, this paper also investigates whether the decreasing impact of decoys on uncertainty levels during decision making is affected by the decoy minimization approach. [ABSTRACT FROM AUTHOR]

Tags in Web 2.0 are becoming another important information source to profile users' interests and preferences to make personalized recommendations. To solve the problem of low information sharing caused by the free-style vocabulary of tags and the long tails of the distribution of tags and items, this paper proposes an approach to integrate the social tags given by users and the item taxonomy with standard vocabulary and hierarchical structure provided by experts to make personalized recommendations. The experimental results show that the proposed approach can effectively improve the information sharing and recommendation accuracy. [ABSTRACT FROM AUTHOR]

This paper aims at studying the utilization of Intelligent Agents for supporting citizens to access e-government services. For this purpose, it proposes a multi-agent system capable of suggesting to the citizens the most interesting services for them; these suggestions are determined by considering both their needs/preferences and the capabilities of the devices used by them. The paper first describes the proposed system and, then, reports various experimental results. Finally, it presents a comparison between the proposed system and other related ones already presented in the literature. [ABSTRACT FROM AUTHOR]

Demographic data regarding users and items exist in most available recommender systems data sets. Still, there has been limited research involving such data. This work sets the foundations for a novel filtering technique which relies on information of that kind. It starts by providing a general, step-by-step description of an approach which combines demographic information with existing filtering algorithms, via a weighted sum, in order to generate more accurate predictions. U-Demog and I-Demog are presented as an application of that general approach specifically on User-based and Item-based Collaborative Filtering. Several experiments involving different settings of the proposed approach support its utility and prove that it shows enough promise in generating predictions of improved quality. [ABSTRACT FROM AUTHOR]

In this paper we present X-Compass, an XML-based agent model for supporting a user in his Web activities. X-Compass is the result of our attempt of synthesizing, in a unique context, important guidelines currently characterizing the research in various Computer Science sectors. Indeed, it constructs and handles a rather rich, even if light, user profile; this latter is exploited for supporting the user in an efficient search of information of his interest; in this way, it behaves as a content-based Recommender System. Moreover, it is particularly suited for constructing multi-agent systems and, therefore, for implementing collaborative filtering recommendation techniques. In addition, since it widely uses XML technology, it is particularly light and capable of operating on various hardware and software platforms. The adoption of XML also facilitates the information exchange among X-Compass agents and, consequently, makes the management and the exploitation of X-Compass based multi-agent systems easier. [ABSTRACT FROM AUTHOR]